Synthesis, characterization and photovoltaic properties of a new thiophene-based double-cable polymer with pendent fullerene group

Article abstract of DOI:10.1016/j.polymer.2012.03.040

In this paper we report the synthesis of a new thiophenic 3-substituted monomer with a plastifying side chain bearing a C₆₀-fullerene moiety at the end. The monomer has been copolymerized with 3-hexylthiophene through the simplest chemical method, namely oxidative polymerization with iron trichloride. The final copolymer, with a high fullerene content (1:1 wt/wt), was fully soluble in common organic solvents and its solubility allowed for its complete physic-chemical characterization. The copolymer thermal behavior was investigated by TGA analysis showing that the incorporation of fullerene has beneficial effects on thermal stability. The fluorescence spectroscopy showed a strong quenching of fluorescence of the copolymer main chains in film, while the AFM analysis confirmed the homogeneous morphology of its films with only very small and uniformly distributed aggregates. The photovoltaic performances of the copolymer and of a film obtained by blending the prepared monomer with poly(3-hexylthiophene) were significantly higher than those obtained using conventional materials and architectures.

Full text of DOI:10.1016/j.polymer.2012.03.040

Polymer 53 (2012) 2134e2145
Contents lists available at SciVerse ScienceDirect
Polymer
journal homepage: www.elsevier.com/locate/polymer
Synthesis, characterization and photovoltaic properties of a new thiophene-based
double-cable polymer with pendent fullerene group
*
Massimiliano Lanzi , Luisa Paganin, Francesco Errani
Dipartimento di Chimica Industriale e dei Materiali, Alma Mater Studiorum-Università di Bologna, Viale del Risorgimento, 4 40136 Bologna, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
In this paper we report the synthesis of a new thiophenic 3-substituted monomer with a plastifying side
chain bearing a C60-fullerene moiety at the end. The monomer has been copolymerized with 3-
hexylthiophene through the simplest chemical method, namely oxidative polymerization with iron
trichloride. The final copolymer, with a high fullerene content (1:1 wt/wt), was fully soluble in common
organic solvents and its solubility allowed for its complete physic-chemical characterization. The
copolymer thermal behavior was investigated by TGA analysis showing that the incorporation of
fullerene has beneficial effects on thermal stability. The fluorescence spectroscopy showed a strong
quenching of fluorescence of the copolymer main chains in film, while the AFM analysis confirmed the
homogeneous morphology of its films with only very small and uniformly distributed aggregates. The
photovoltaic performances of the copolymer and of a film obtained by blending the prepared monomer
with poly(3-hexylthiophene) were significantly higher than those obtained using conventional materials
and architectures.
Received 19 November 2011
Received in revised form
18 March 2012
Accepted 21 March 2012
Available online 29 March 2012
Keywords:
Polythiophene
Bulk heterojunction solar cells
Fullerene
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
to its easy synthesis, good solubility and thermal stability, low
band-gap and high charge-carrier mobility [8,9]. Moreover, the
Conjugated polymers (CP)s have received great interest in the
recent years as they offer a simple, innovative and versatile route
towards the development of cost-effective, lightweight and flexible
organic photovoltaic (OPV) solar cells [1e4]. The most studied and
applied structure is the bulk heterojunction (BHJ) architecture,
where the photoactive layer consists of an interpenetrating
electro-optical properties of this polymer are strongly influenced
by its film morphology, strongly related to the conformational
order of the polymeric backbone, and easily controllable by post-
production treatments such as thermal annealing, solvent varia-
tion or the application of electrical fields. In particular, the post-
treatment of the blend by thermal annealing has proved to be
very effective in improving the power conversion efficiency (PCE) of
the final device [10,11] acting not only on P3HT conformational
order but also on the blend morphology. In fact, the donor and
acceptor molecules are generally not very compatible and tend to
segregate one from the other causing phase separation [12].
Uncontrolled macrophase separation has a negative effect on
cell performance, leading to the formation of domains of the two
components that are larger than the exciton diffusion lengths,
hindering photoinduced charge separation and transportation in
the active blend [13].
The higher efficiency obtained after the thermal treatment of
the devices can be ascribed to the enhanced hole mobility in the
P3HT phase determined by thermal induced crystallization of the
polymer and by the better charge-carrier generation enhanced by
the gain of a nanoscale separation between the two components.
Some recent works clearly demonstrate that thermal annealing of
the active blend, executed before or also after the deposition of the
network (blend) of a
p-conjugated polymer (electron donor) and
a soluble fullerene derivative (electron-acceptor). The low effi-
ciency of the polymeric solar cells, if compared with the silicon
solar cells, as well as their reduced thermal stability in time are still
the main limiting factors for their wide-spread commercialization
[5]. Even if the bulk heterojunction structure can allow an ultrafast
charge separation from the conjugated polymer to the fullerene [6],
a high interfacial area between the two components, the formation
of a percolation path for charge transport [7] and the control of
morphology in the composite active layer are of crucial importance
for the performance of the device.
Poly(3-hexylthiophene) (P3HT) has emerged as a valuable
conjugated polymer for the fabrication polymeric solar cells thanks
* Corresponding author. Tel.: þ39 051 2093689; fax: þ39 051 2093669.
E-mail address: massimiliano.lanzi@unibo.it (M. Lanzi).
0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.polymer.2012.03.040

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2135
metallic cathode, is a fundamental prerequisite for obtaining highly
efficient polymeric solar cells [14,15].
measurements. Finally, the current density/voltage (J/V) character-
istics of the obtained polymer solar cells have been recorded under
one-sun illumination to determine PCE when employing the newly
synthesized photoactive monomeric and polymeric materials.
In addition to the physical (thermal) approach, some chemical
methods can be used to control morphology within the photoactive
layer, with the aim of minimizing donoreacceptor macrophase
separation. One of these methods involves the synthesis of p-type
conjugated backbones bearing directly grafted fullerene moieties.
This way is particularly interesting and chemically neat since
a single polymeric material could possess the capability of both
electron and hole transport (double-cable polymer), thus leading to
faster inter- and intra-chain transfer of the positive holes and also
to an easier hopping of the electrons among the pendent fullerenes.
Moreover, the interfacial area between the donor and acceptor
groups can be maximized and the clustering phenomena avoided
[16].
In this paper, we report both the synthesis of a new thiophenic
monomer bearing the fullerene moiety at the end of an alkylic side
chain and the preparation of its homopolymer and copolymers
with a different content of 3-hexylthiophene used as a solubility
helper (Fig. 1). The soluble products have been characterized by FT-
IR, NMR, GPC, UVeVis and fluorescence emission spectroscopy in
order to assess the polymeric structure and investigate their elec-
tronic properties.
2. Experimental
2.1. Materials
All reagents were purchased from SigmaeAldrich Chemical Co.
and used without further purification unless expressly indicated
otherwise. All solvents employed (HPLC grade) were purified by
normal procedures, stored under molecular sieves and handled in
a moisture-free atmosphere.
Poly(3-hexylthiophene) (P3HT) was synthesized starting from
3-hexylthiophene (T6H) bought from SigmaeAldrich (CAS N. 1693-
86-3, Product N. 399051) using the McCullough procedure [17]:
regioregularity in HT dyads 98%, Mn 45 kDa, PDI 1.4.
2.2. Measurements
FT-IR of monomers and polymers were recorded on a Perkin
Elmer Spectrum One Spectrophotometer by drop casting pure
samples or their solutions in spectroquality solvents onto Ge disks
The molecular arrangement and morphology of the blends
prepared using either the synthesized donor-p-acceptor double-
or by making KBr pressed pellets of the powder samples. 1H and 13
C
cable polymers or the synthesized monomer (T6F) mixed with
P3HT have been studied by means of optical spectroscopy and AFM
NMR spectra were recorded on a Varian Gemini 300 (300 MHz)
instrument using tetramethylsilane (TMS) as an internal reference.
UVeVis spectra were carried out on polymer films cast from chlo-
robenzene (CB) solutions on quartz slides by means of a BLE 3000
spin coater operating at 1750 rpm for 5 min, and measured using
a Perkin Elmer Lambda 19 spectrophotometer. Fluorescence spectra
of polymeric blends were measured by means of an Edinburgh FLSP
920 spectrofluorimeter using thin films on quartz slides. Molecular
weights and polydispersity of the polymers were determined by
GPC analysis, with a polystyrene standards calibration, using THF
solutions on a PL Laboratory UVis 200 instrument, equipped with
a Phenogel mixed bed column. A DSC TA Instruments 2920 was
used for the thermal analysis of the polymers by varying the
temperature from ꢀ50 to 200 ꢁC at a rate of 10 ꢁC/min in a nitrogen
atmosphere. A TGA TA Instruments 2050, operating either in an
inert or oxidizing atmosphere, was used to determine the decom-
position temperatures for the polymers by heating samples from 40
to 900 ꢁC at a heating scan rate of 20 ꢁC/min.
CH3
N
(CH2)6
O
S
T6F
(CH2)6
H
S
X-ray diffraction data of polymers (powder and thin films cast
from CB, thickness about 100 nm) were collected at room
FeCl3
CH3NO2/CHCl3
temperature using CuK
Bragg-Brentano Powder Diffractometer Philips PW1050/81-
PW1710 equipped with a graphite monochromator in the dif-
fracted beam. A 2
range between 2.0 and 90.0ꢁ was scanned by
a
source radiation (
l
¼ 1.5406 Å) and
a
q
881 steps of 0.1ꢁ each with a counting time of 15 s for each step.
Polymer solar cells were fabricated using the basic ITO (100 nm)/
PEDOT:PSS (50 nm)/Active layer (about 150 nm)/Al (100 nm) archi-
tecture. Theactivelayerwaseitherathinfilmofdouble-cablepolymer
CH3
N
(CH2)6
m
O
(CH2)6
H
or a blend of P3HT with different electron-acceptor molecules (C60
-
fullerene, PCBM or the synthesized monomer bearing the fullerene
moiety in the side chain) in 1:1 weight ratio. The commercial ITO-
coated glass, with a thickness of the InO2/SnO2 layer of about
S
S
n
100 nm and a sheet resistance lower than 15
U (Merck), was etched on
one side with acid, cleaned using acetone in an ultrasonic bath, and
then subjected to an RCA-SC1 (20:1:4 volume ratio of H2O, NH4OH
and H2O2 for 20 min at 60 ꢁC) treatment. A film of poly(3,4-
ethylenedioxythiophene):polystyrene sulfonate (PEDOT:PSS, thick-
ness about 50 nm) was coated onto the ITO substrate by means of the
doctor blading technique using commercial aqueous solution (Bay-
tron PSS) diluted 1:1 (v/v) with isopropanol, and subsequently
P1 n=0, m=1
CP1 n=0.7, m=0.3
CP2 n=0.5, m=0.5
Fig. 1. Synthesized monomer and polymers.

2136
M. Lanzi et al. / Polymer 53 (2012) 2134e2145
annealed in a vacuum for 2 h at 120 ꢁC. The active layer was then cast
using a BLE 3000 spin coater operating at 1500 rpm using the CB
solution of donor and acceptor, previously filtered through a 0.45 mm
a mixture of 0.30 g (1.78 mmol) of 3-hexylthiophene (T6H) and
0.79 g (0.76 mmol) of T6F in 26 ml of anhydrous CHCl3. After the
addition was completed, the reaction mixture was stirred for 1 h at
room temperature and added to 30 ml of freshly distilled THF. The
mixture was then poured into 125 ml of a 5% methanolic solution of
HCl and the resulting precipitate filtered off, washed with further
methanol and dried under reduced pressure at room temperature,
thus giving the crude polymer as a black powder. The resulting
material was then re-dissolved in 300 ml of toluene, slowly added
with 150 ml of aqueous 2% HCl and the resulting mixture stirred
overnight. The organic phase was washed with water to neutrality,
dried with MgSO4 and concentrated under reduced pressure giving
0.72 g (66% yield) of CP1.
PTFE filter, and dried under a gentle nitrogen flow (final thickness
about 150 nm). The device was then annealed in a vacuum for 30 min
at 150 ꢁC and the Aluminum top electrode was deposited by thermal
evaporation in an ultra-high vacuum (10ꢀ6 mbar) using an Edwards
E306A coating system, giving a metal thickness of 100 nm. The
photovoltaic performances of the samples were measured by means
of aKeithley 2401sourcemeasurementunitinair. AM1.5illumination
(100 mW/cm2) was provided by an Abet Technologies LS 150 Xenon
Arc Lamp Source calibrated with an ILT 1400-BL radiometer
photometer. Thickness and homogeneity of the different active layers
was determined using a Burleigh Vista AFM working in a non-contact
1H NMR (CDCl3, ppm)
d: 7.75 (m); 6.98 (bm); 4.95 (m); 4.82
tapping mode on a 10
m
m ꢂ 10
m
m scanning area and by means of
(bm); 4.20 (bm); 3.90 (bm); 2.85 (m); 2.60 (m); 1.80 (bm), 1.35
(bm), 0.90 (m).
a SEM ZEISS EVO 50 EP electronic microscope at 10,000ꢂ magnifi-
cation. Phase images were acquired using a high resolution (10 Å)
head equipped with a silicon-nitride tip.
13C NMR (CDCl3, ppm)
d: 160.0, 156.3, 154.0, 153.4, 147.3, 146.7,
146.5, 146.3, 146.1, 145.9, 145.7,145.5, 145.3,145.1,144.7,144.4,143.1,
142.9, 142.7, 142.5, 142.3, 142.1, 142.0, 141.8, 141.6, 141.4, 140.1, 139.8,
139.5, 139.2, 137.4, 136.8, 136.6, 136.4, 135.9, 135.8, 128.8, 125.8,
120.6, 115.3, 84.2, 70.8, 68.5, 66.7, 40.7, 31.3, 30.6, 29.9, 29.6, 26.4,
22.8, 14.2.
2.3. Monomer synthesis
2.3.1. 3-[6-(1-oxy-4-formylphenyl)hexyl]thiophene] (T6BA)
2.00 g (8.09 mmol) of 3-(6-bromohexyl)thiophene prepared
according to Ref. [18],1.46 g (11.95 mmol) of 4-hydroxybenzaldehyde
and 6.04 g (4.37 mmol) of K2CO3 were added to 16 ml of N,N-
dimethylformamide (DMF). The mixture was reacted for 8 h at
80 ꢁC and then poured into 150 ml of distilled water and extracted
several times with n-pentane. The collected organic phases were
washed with distilled water to neutrality, dried with MgSO4 and
concentrated, yielding 2.2 g of crude product which was subjected to
column chromatography purification (SiO2/n-heptane:ethyl acetate
7:3) giving 1.81 g (6.27 mmol, 78% yield) of pure T6BA.
FT-IR (Ge): 3050; 3028; 2953; 2923; 2851; 1608, 1581; 1510;
1460; 1428; 1376; 1243; 1181; 1173; 839; 574, 525 cmꢀ1
.
Elemental analysis: Calcd. for [(C79H23NOS)0.3(C10H14S)0.7]n: C
86.43; H 3.95; N 0.97; O 1.13; S 7.52; Found: C 86.91; H 3.75; N 0.99;
O 1.11; S 7.24.
2.3.4. Poly{3-[6-(1-oxy-4-N-methylfulleropyrrolidinephenyl)hexyl]
thiophene} (P1)
The adopted procedure was the same as for CP1 but using only
the monomer T6F. The final polymer was insoluble in common
organic solvents and this prevented its purification using the
toluene/aqueous 2% HCl mixture (91% yield).
1H NMR (CDCl3, ppm)
d
: 9.90(s,1H, H16); 7.82(m, 2H, H13 þ H14);
7.25 (m, 1H, H5); 6.95 (m, 4H, H2 þ H4 þ H12 þ H15); 4.00 (t, 2H,
H11); 2.62 (t, 2H, H6); 2.00e1.20 (m, 8H, H7 þ H8 þ H9 þ H10).
FT-IR (KBr): 3051; 3026; 2955; 2922; 2850; 1607, 1581; 1510;
13C NMR (CDCl3, ppm)
d
: 191.4; 164.9; 143.6; 132.6; 130.4;
1465; 1428; 1377; 1242; 1180; 1173; 837; 766, 660; 574, 522 cmꢀ1
.
128.9; 125.8; 120.5; 115.4; 69.0; 31.1; 30.8; 29.6; 29.4; 26.4.
FT-IR (KBr): 3101; 3050; 3029; 2932; 2856; 2736; 1690; 1601,
Elemental analysis: Calcd. For (C79H23NOS)n: C 91.76; H 2.24; N
1.35; O 1.55; S 3.10; Found: C 92.12; H 2.18; N 1.41; O 1.58; S 2.71.
1577; 1508; 1466; 1385; 1285; 1159; 774, 685 cmꢀ1
.
2.3.5. Poly{3-hexylthiophene-co-3-[6-(1-oxy-4-N-
2.3.2. 3-[6-(1-oxy-4-N-methylfulleropyrrolidinephenyl)hexyl]
thiophene (T6F)
methylfulleropyrrolidinephenyl)hexyl]thiophene} (CP2)
The adopted procedure was the same as for CP1 but using a T6F/
T6H 1:1 feed molar ratio. The final polymer was insoluble in
common organic solvents and this prevented its purification using
the toluene/aqueous 2% HCl mixture (88% yield).
1.00 g (3.47 mmol) of T6BA, 1.23 g (13.8 mmol) of N-methyl-
glycine (sarcosine) and 5.00 g (6.94 mmol) of C60-fullerene were
added to 500 ml of 1,2-dichlorobenzene. The mixture was refluxed
for 18 h under nitrogen. After evaporation of the solvent at reduced
pressure, the crude product was purified by column chromatog-
raphy (SiO2; toluene:n-heptane 1:1) leading to 2.51 g (2.43 mmol,
70% yield) of pure T6F.
FT-IR (KBr): 3050; 3027; 2955; 2925; 2851; 1608, 1582; 1511;
1465; 1428; 1373; 1241; 1181; 1173; 839; 764, 661; 573, 522 cmꢀ1
.
Elemental analysis: Calcd. for [(C79H23NOS)0.5(C10H14S)0.5]n: C
88.66; H 3.23; N 1.14; O 1.30; S 5.67; Found: C 88.23; H 3.48; N 1.16;
O 1.39; S 5.74.
1H NMR (CDCl3, ppm)
d
: 7.71 (d, 2H, H13 þ H14); 7.20 (m, 1H,
H5); 6.91 (m, 4H, H2 þ H4 þ H12 þ H15); 4.98 (d, 1H, H17); 4.82 (s,
1H, H16); 4.22 (d, 1H, H17); 3.91 (t, 2H, H11); 2.78 (s, 3H, H18); 2.60
(t, 2H, H6); 1.82e1.30 (m, 8H, H7 þ H8 þ H9 þ H10).
3. Results and discussion
13C NMR (CDCl3, ppm)
d
: 159.8, 156.6, 154.2, 153.7, 147.9, 146.9,
Since uniformity and homogeneity of
C60 thin films are fundamental prerequisites for obtaining high
performance optoelectronic devices, the design of soluble
p-conjugated polymers/
146.6, 146.3, 146.1, 145.9, 145.7, 145.5, 145.1, 144.9, 144.6, 144.3,
143.5, 142.9, 142.7, 142.5, 141.9, 141.5, 140.1, 139.7, 139.4, 139.1, 137.7,
136.8, 136.6, 136.4, 135.9, 135.7, 131.1, 128.9, 125.8, 120.5, 115.2, 83.9,
70.6, 68.5, 66.6, 40.7, 31.1, 30.8, 29.9, 29.7, 26.6, 23.0.
a
fullerene-functionalized thiophenic derivative was pursued. In fact,
T6F synthesis was followed in order to have good control over the
morphology and phase separation of the interpenetrating networks
of the photoactive blends in the organic solar cells by the simple
substitution of fullerene or PCBM with the newly synthesized
electron-acceptor molecule. Moreover, T6F is a thiophenic deriva-
FT-IR (KBr): 3101; 3050; 3030; 2924; 2850; 2776; 1609, 1581;
1509; 1461; 1427; 1331; 1245; 1173; 768, 661; 574, 526 cmꢀ1
.
2.3.3. Poly{3-hexylthiophene-co-3-[6-(1-oxy-4-N-
methylfulleropyrrolidinephenyl)hexyl]thiophene} (CP1)
1.50 g (9.26 mmol) of iron trichloride in 10 ml of nitromethane
was dropped in 45 min, under argon and at room temperature, into
tive that is not substituted in the
therefore, it can be directly used as a monomer or comonomer to
synthesize conjugated polymer bearing covalently linked
a-positions of the aromatic ring:
a

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2137
Table 1
fullerene moieties in the side chains, thus being a viable candidate
Main IR absorption bands of monomers and polymers.
for the fabrication of stable bicontinuous networks with a “double-
cable” structure [19,20]. The presence of an alkyl spacer between
the polythiophene backbone and the fullerene group should also
offer many advantages, such as:
T6BA T6F
CP1
P1
CP2
Assignment
3101 3101
3050 3050 3050 3051
3029 3030 3028 3026
e
e
e
ns CeH
ns CeH
a
b
-thiophene [28]
-thiophene [29e31]
3050
3027
2955
2925
2851
emb
e
n
n
n
n
n
n
n
n
CeH phenyl ring [32]
as eCH3 [33]
as eCH2e [30,34]
s eCH2e [30,34]
eN eCH3 [35]
H eCO [35]
C]O [36]
C]C 1,4-disubstituted
e
e
2953 2955
- better solubility of the polymer, acting as an internal plastifying
group;
- easier polymerization of the monomer, since the spacer insu-
lates the thiophenic unit from the electron-withdrawing
fullerene group;
- higher photovoltaic effect, according to Yassar et al. [21] which
have shown that in these systems no charge transfer occurs in
the ground state.
2932 2924 2923 2922
2856 2850 2851 2850
e
2776 emb emb
2736
1690
e
e
e
e
e
e
e
1601, 1609, 1608, 1607,
1577 1581 1581 1581
1508 1509 1510 1510
1466 1461 1460 1465
1608,
1582
1511
1465
1428
e
phenyl ring [35,37,38]
nas C]C thiophene [29e31]
ns C]C thiophene [29e31]
Fullerene [35,39,40]
e
1427 1428 1428
1385
e
e
e
e
d
H eC formyl [36]
Methyl deformation [34]
N eC amine [37]
Scheme 1 outlines the synthetic strategy adopted to obtain T6F
monomer while Fig. 1 indicates the method used for the prepara-
tion of T6F homopolymer and copolymers. The chosen pathway is
in fact straightforward and effective and leads to the desired
double-cable polymers with good yields in just a few simple steps.
3-(6-bromohexyl)thiophene (T6Br) was synthesized according
to Ref. [18] and subjected to an SN2 reaction with p-hydrox-
ybenzaldehyde in DMF in presence of K2CO3, leading to 3-[6-(1-
oxy-4-formylphenyl)hexyl]thiophene (T6BA) with 78% yield. This
aldehydic derivative was then condensed with sarcosine in the
presence of C60-fullerene [22], giving the final monomer 3-[6-(1-
oxy-4-N-methylfulleropyrrolidinephenyl)hexyl]thiophene (T6F) in
70% isolated yield. The spectroscopic data (see FT-IR in Table 1 and
Fig. 2 and NMR in the Experimental section which reports the main
assignments) were consistent with the expected structures.
The (co)polymers were synthesized by reacting the monomer(s)
with iron trichloride in the CH3NO2/CHCl3 mixture, with a proce-
dure that allowed the precipitation of the oxidizing agent in
a highly active microcrystalline form [23,24]. In fact, the slow
addition of FeCl3 facilitates control of the Fe(III)/Fe(II) couple
oxidation potential during the polymerization reaction, while
minimizing the insoluble fraction content.
The use of this polymerization procedure together with
employment of the alkylic plastifying side chains was however
insufficient to obtain a soluble homopolymer. In fact, the homo-
polymerization of T6F leads to a black powder which was insoluble
in common organic solvents but which had an elemental analysis
that complied with the expected structure. The absence of oxidant
residues was also confirmed by FT-IR analysis (absence of the
intense doping bands usually observed in the 1350e1000 cmꢀ1
range [25]), thus suggesting that the poor solubility of P1 should
emb 1376 1377
1331 emb emb
1373
emb
1241
1173
839
e
n
1258 1245 1243 1242
nas CeOeC [41]
Fullerene [35,39,40]
e
1173 1173 1173
e
e
839
837
g
CeH 2,3,5-trisubstituted
thiophene [29e31]
CeH 1,4-disubstituted
phenyl ring [42]
þ
g
774
e
768
e
e
e
e
g
g
CeH 3-substituted thiophene [37]
phenyl ring [42]
emb 766,660 764,661
e
574, 574, 574, 522 573, 522 Fullerene [35,39,40]
526 525
emb ¼ embedded.
be directly ascribed to the linkage of fullerene moieties to the
conjugated backbone, as already reported in the literature [26].
Since a high fullerene content is necessary for an appreciable
external quantum efficiency of photodiodes and high photocur-
rents [27], we tried to copolymerize T6H and T6F in a 1:1 M ratio.
The final copolymer was already insoluble but its composition
reflected the monomers feed ratio, thus indicating that the T6H and
T6F reactivity towards the oxidative polymerization with FeCl3 was
almost the same.
FT-IR bands of monomers and polymers as well as their
assignments are reported in Table 1.
The copolymer prepared using a 30% mol of T6F and the same
polymerization technique was instead completely soluble in
common organic solvents and was recovered, after a careful puri-
fication procedure, as a fine black powder with a metallic luster and
with a good yield (66%). The comparison between its IR spectrum
and that of the monomer T6F (Fig. 2) shows both the complete
disappearance of the band at 3101 cmꢀ1 (symmetric stretching of
CeH in
a to the thiophenic ring) and the appearance of the signal at
839 cmꢀ1 (bending out of plane of CeH in 2,3,5-trisubstituted
thiophene) confirming the formation of a polythiophene with
high molecular weight; at the same time, the presence of the
15 14
12 13
O
H
O
10
absorptions at 1428, 1173, 574 and 525 cmꢀ1, ascribable to the C60
-
8
11
16
O
H
6
(CH2)6 Br
9
4
3
fullerene, as well as the bands at 3028, 1608, 1581, 1243, 1181,
directly deriving from T6F monomer structure, confirms the
formation of the expected copolymer. The spectrum of the homo-
polymer P1 is qualitatively similar to that of the copolymer CP1
except for the absorption at about 3350 cmꢀ1 ascribable to the
presence of traces of water in the KBr powder used to make the
pressed pellet for the analysis. In fact, the insolubility of P1 in
common organic solvents prevented its analysis as a thin film on Ge
disk. The IR spectrum of CP2, not reported here, is almost identical
to the spectrum of the homopolymer.
HO
7
2
5
S
K2CO3
S
18
CH3
1
N,N-DMF
T6BA
T6Br
14
15
N
16
17
O
10
CH3 NH CH2 COOH
8
12
13
11
6
9
3
4
5
7
2
S
1
Cl
Cl
T6F
Comparison of the 1H NMR spectra of CP1 and of the como-
nomer T6F, reported in Fig. 3, clearly shows the structure of the
synthesized copolymer; in particular, there are two distinct signals
Scheme 1. Synthesis of monomer T6F.

2138
M. Lanzi et al. / Polymer 53 (2012) 2134e2145
Fig. 3. 1H NMR of monomer T6F (top) and copolymer CP1 (bottom).
at 4.20 and 4.95 ppm, which are attributable to the two germinal
protons of the methylenes in the pyrrolidine ring, as well as the
signal at 4.82 ppm which is ascribable to the remaining proton of
the same ring. Unfortunately, the signal due to the N-methyl
protons of the fulleropyrrolidine group, at 2.78 ppm in T6F, is
embedded with the signal deriving from the methylenic protons a-
to the thiophene rings in HT dyads, thus making the correct
determination of both the regioregularity and the composition of
CP1 more difficult. The composition of this copolymer was then
determined by comparing the integral of the signal of the eCH3
group belonging to the T6H monomer at 0.90 ppm, with the inte-
gral of the peak at 3.90 ppm, relative to the methylenes a- to the
phenoxy group, which is present only in T6F monomer. The HT
dyads CP1 percentage can then be obtained by subtracting the
contribute of the NeCH3 protons to the integral of the peak at
2.85 ppm and calculating the ratio of the resulting value with the
integral of the signal at 2.60 ppm, which is ascribable to the non-HT
linked dyads.
CP1 has a 30% molar content of T6F and 70% HT percentage. Its
composition is in agreement both with the two comonomers feed
ratio and with the composition determined by elemental analysis.
Its molecular weight characteristics were easily determined by
means of GPC analysis thanks to its high solubility, showing: Mn
53 kDa, Mw/Mn 1.55 and a DPn of about 124 repeating units. The
Fig. 2. FT-IR spectra of T6BA, T6F, CP1 and P1.

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2139
high fullerene content (1:1 wt/wt, relative to the fullerene net
amount in the copolymer) as well as its high molecular weight and
good solubility are indeed promising characteristics for the appli-
cation of CP1 for assembling OPV solar cells and justify a more
detailed analysis of this copolymer.
the presence of the fullerene group, a sterical demanding bulky
molecule, and is confirmed by the UVeVis analysis of CP1 films.
Fig. 5 shows the UVeVis absorption spectra of the reference
polymer P3HT, of the copolymer CP1 and of a 1:1 (wt/wt) blend of
P3HT and T6F in chlorobenzene (CB) solution and in film cast from
CB solutions before and after the annealing procedure.
The thermal properties of the soluble double-cable copolymer
CP1 were analyzed by TGA and compared with those of the refer-
ence polymer P3HT. The TGA analysis was carried out in nitrogen
and in air with a heating rate of 20 ꢁC/min. In Fig. 4 the thermo-
grams of the two polymers in both the examined environments are
reported. All samples exhibit one-step weight loss processes with
onset temperatures around 320 ꢁC (P3HT) and 410 ꢁC (CP1) in air
and 410 ꢁC (P3HT) and 490 ꢁC (CP1) in nitrogen. The double-cable
polymer exhibits a thermal stability that is notably higher than
the reference polymer devoid of the fullerene moiety. CP1 residual
weight observed in the non-oxidizing atmosphere is roughly in
agreement with its chemical structure, taking into account its
fullerene content and the presence of a non volatile residue
deriving from its polythiophenic backbone, as observed in P3HT
sample. The absence, in CP1 thermograms, of any inflection point is
in agreement with a random copolymer structure [43].
The DSC analysis of CP1 was carried out in a nitrogen atmo-
sphere with a heating rate of 10 ꢁC/min. The absence of any evident
endo- or exothermic peak denotes the absence of long-range order
in the synthesized copolymer: a fact which is probably ascribable to
Fig. 4. TGA thermograms of the copolymer CP1 and of the reference polymer P3HT in
nitrogen (top) and in air (bottom).
Fig. 5. UVeVis spectra of polymers in solution and film.

2140
M. Lanzi et al. / Polymer 53 (2012) 2134e2145
In the film before the annealing procedure P3HT exhibits a lmax
at 522 nm while the blend shows two absorptions: the first one at
336 nm, ascribable to the C60 moiety [44,45] and the second, at
454 nm, ascribable to the polyconjugated backbone (Table 2). The
blue shift of the lmax of the polythiophenic
p-p* system suggests
a perturbation of the molecular arrangement of the P3HT backbone
by the copresence of the fulleropyrrolidinic derivative T6F. This
phenomenon is more evident in the case of CP1, where the two
absorptions are embedded and undistinguishable from each other.
The fullerene, although linked to the polymeric backbone by
a flexible spacer, leads to a steric interaction among the side chains;
this interaction is able to prevent a high planarity degree of the
conjugated system, influencing the mean conjugation length and
the absorption wavelength in the solid state. For P3HT and for the
blend, after the annealing step, the lmax of the polyconjugated
component is red-shifted with respect to the non-annealed film
(from 454 to 460 nm and from 522 to 536 nm for the blend and
P3HT, respectively) suggesting a more extended conjugation length
in the annealed films. Moreover, an evident structure appears in
P3HT spectra (after and before annealing) around 600 nm, indi-
cating a high level of molecular ordering (p-stacking) [46,47]. This
structure is completely lost in CP1 and blend film spectra, where an
inhomogeneous broadening of the polythiophenic component and
a blue shift of the absorption maxima (with respect to P3HT) take
place, indicating that the interactions among the polymer chains
were disrupted by the presence of the C60 molecules [48]. A further
confirmation arises from the analysis of the spectra of CP1 and of
the blend in solution, which are very similar to those recorded in
the film state.
A high fullerene content in the blend is necessary for the
photovoltaic effect of BHJ solar cells, since the percolation threshold
for the electronic transport has to be overcome, but a high content
of electron-acceptor molecules seems to be unfavorable not only
for the solubility of the two components of the photoactive blend,
but also for the planarity of the polyconjugated backbone. This is
particularly true when the fullerene is directly linked to the poly-
mer rather than in the case of the blend, where the two compo-
nents are simply mixed.
Fig. 6 shows the photoluminescence (PL) spectra of the reference
polymer P3HT, of the double-cable copolymer CP1 and of the blend
P3HT-T6F in film cast from CB solutions on quartz slides before and
after annealing for 15 min in an inert atmosphere at 150 ꢁC. From the
comparison of P3HT spectra before and after annealing, it is evident
that the polymer undergoes both a bathochromic shift of the
maximum absorption wavelength (from 730 to 745 nm) and a weak
hyperchromic effect thanks to the formation of more ordered and
conjugated conformers. The blend and CP1 in particular show a high
quenching of the fluorescence intensity if compared with P3HT, and
this is ascribable to the intramolecular photoinduced electron-
transfer from the polyconjugated backbone to the fullerene moiety.
The charge transfer is probably lower for the blend because the
fulleropyrrolidinic derivative is not directly linked to the conjugated
backbone. In any case the effectiveness of the charge transfer for
randomly grafted fullerene-derivatized double-cable polymers has
already been described in the recent literature [43]. It is also evident
from Fig. 6 that the thermal treatments have beneficial effects on the
Fig. 6. Photoluminescence (PL) spectra of the polymers and blend in film after (top)
and before (bottom) the annealing procedure (see text).
photoinduced charge transfer, as evidenced by the stronger
quenching of the fluorescence intensity [12,49].
Figs. 7 and 8 show the XRD patterns of P3HT, CP1 and of the
blend P3HT-T6F in powder and in film on quartz slides (cast from
CB solutions, average thickness 150 nm) before and after the
annealing procedure (15 min in inert atmosphere at 150 ꢁC) while
Tables 3 and 4 show their structural data.
The powder diffraction profiles of the examined polymers
exhibit the (100) reflection and, for P3HT and the blend, its higher
order reflections, corresponding to the side chain lamellar stacking
distance. Another reflection is also clearly evident at about
Table 2
UVeVis absorption maxima (lmax) of the polymers in solution and film.
2
q
¼ 23ꢁ, corresponding to the
p-p stacking of the thiophene ring
in the main chain. The positive effect of the annealing is quite
evident, determining the sharpening of the low-angle diffraction
peaks and a better appearance of the high-angle peak. On the other
hand, the presence of the fullerene derivative increases the P3HT
chains disorder, as evidenced by the lower intensity of the peaks
Sample
Chlorobenzene
solution (nm)
Film before
annealing (nm)
Film after
annealing (nm)
P3HT
Blend
CP1
448
319, 439
328
522
336, 454
342
536
336, 460
342

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2141
Fig. 7. X-rays diffractograms of the polymers in powder before (left) and after (right) the annealing procedure.
and by the absence of high-order reflections in the diffraction
profile of the copolymer.
sharper and more intense than the pristine one, denoting the
effectiveness of the annealing procedure in enhancing the struc-
tural order of the conjugated polymer. XRD analysis then suggests
that P3HT in thin film has an edge-on structure with a preferential
orientation of the (100) axis normal to the film [52].
The XRD analysis results of the blend and CP1 samples are quite
different from those recorded for P3HT. In fact, in both cases, the
interlayer d1 spacings are slightly larger than for P3HT, indicating
that the presence of the C60 moiety can reduce the order and degree
of crystallinity of the polythiophenic chains by creating a persistent
sterical hindrance capable of perturbing the lamellar stacking of the
Fig. 8 exhibits XRD patterns of the polymers in film. P3HT shows
a small angle reflection at 2
q
¼ 5.42ꢁ (d1 ¼ 16.3 Å) which corre-
sponds to the first order reflection ascribable to the distance
between polythiophenic chains lying on the same plane. The two
other peaks at 10.9 and 16.4ꢁ are also detected and correspond to
(200) and (300) reflections induced by the multiple X-ray reflec-
tions by the film. The obtained interlayer distance agrees with the
data reported in literature for poly(3-hexylthiophene) in film
[50,51]. The annealed sample shows a peak that is significantly

2142
M. Lanzi et al. / Polymer 53 (2012) 2134e2145
Fig. 8. X-rays diffractograms of the polymers in film before (left) and after (right) the annealing procedure.
Table 3
Structural parameters of the synthesized polymers in powder.
Table 4
Structural parameters of the synthesized polymers in film.
Sample
Low-angle
diffractions
(degrees)
High-angle On plane
diffractions PT chains
Planes
stacking
Sample
Low-angle diffractions
(degrees)
On plane PT
chains distances (Å)
(degrees)
distances (Å) distances (Å)
P3HT pristine
Blend pristine
CP1 pristine
P3HT annealed 5.49; 10.9; 16.1 22.8
Blend annealed 5.46; 10.7
CP1 annealed 5.30
5.50; 11.1; 16.5 23.0
16.1
16.1
16.5
16.1
16.2
16.7
3.94
3.96
4.00
3.98
3.98
4.02
P3HT pristine
Blend pristine
CP1 pristine
P3HT annealed
Blend annealed
CP1 annealed
5.42; 10.9; 16.4
5.38
5.20
5.45; 10.9; 16.6
5.37; 10.8
5.18
16.3
16.4
17.0
16.2
16.5
17.1
5.48; 10.8
5.35
22.9
22.6
22.8
22.5

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2143
Fig. 9. AFM images of CP1 and of the prepared blends (tapping mode, scale in nm; left: height modulated images, right: phase modulated images).

2144
M. Lanzi et al. / Polymer 53 (2012) 2134e2145
polymer main chains as also shown by the lower intensities of the
crystallization peaks. This perturbation seems to depend on the
nature of the fullerenic derivative (i.e. blended or linked to the
polymer) and consequently, the annealing procedure does not have
the same effect on all the examined samples, being more effective
with those polymer chains that have a higher free volumes.
The efficiency of the polymeric solar cells prepared according to
the method reported in the experimental section was examined
under AM 1.5 (100 mW cmꢀ2) illumination. Devices were not
optimized for their photovoltaic performances since the results
were finalized to compare their efficiency on a standard prepara-
tion platform, while favoring data reproducibility over the overall
efficiency.
Best results were obtained using the P3HT/T6F system as the
active blend. This is probably due to the high solubility of these two
components in the solvent used for film deposition, thus leading to
a very uniform distribution of the electron donor/electron-acceptor
moieties on nanoscale phase separation. Also CP1 gave a photo-
voltaic efficiency that was higher than the reference blends thanks
to its high fullerene content that probably overcame the percola-
tion threshold for the electronic transport in BHJ solar cells [60], as
well as to the electron-acceptor groups directly grafted on its
polymeric backbone.
Moreover, the broad scattering around 2
q
¼ 22ꢁ is mainly ascribable
to the disordered aggregation of the alkyl chains [53e55], pre-
venting the regular face-to-face packing of the
p-conjugated
groups, which usually gives rise to strong reflections with spacing
in the 3.5e3.9 Å [56e58]. Even if these reflections generally overlap
with the broad scattering, they are not observed in the films of the
studied polymers, indicating that the main chains as well as the
alkyl chains are fairly disordered. This fact can be ascribed not only
to the presence of the fullerenic derivative but also to the annealing
conditions and/or to the solvent used to prepare the films and
suggests that some improvements should be made.
Fig. 9 shows the surface morphology of CP1, of the T6F/PT6H
blend and of two reference blends made with P3HT and C60 or
PCBM 1:1 wt/wt, usually used for the fabrication of BHJ solar cells.
The films were prepared by casting CB solutions onto ITO glasses
that had been previously cleaned with the same procedure adopted
for the preparation of the solar cells. After the deposition of the
polymeric layer, the samples were thermally annealed for 15 min at
150 ꢁC under vacuum. The images were obtained using an AFM in
a non-contact (tapping) mode. The surfaces of the reference blend
(i.e. P3HT/PCBM) and, in particular, of the examined blend (P3HT/
T6F) have a very smooth surface if compared with those of CP1 and
P3HT/C60. The latter has the higher surface roughness (average
roughness: 2.2 nm) while the P3HT/T6F blend has both the
smoother surface (average roughness: 0.5 nm) and the more
regular texture.
In the case of the reference blend P3HT/PCBM the annealing
procedure leads to the formation of wide and well defined domains
of the acceptor but the substitution of PCBM with T6F allows better
control of the phase separation, yielding to very small aggregates
with a highly homogeneous size distribution. The high interfacial
area of the electron-acceptor domains together with their uniform
distribution in the blend should be beneficial for the photovoltaic
properties, leading to a better control of the separation of the
components of the active blends on a nanoscale domain.
AFM tapping mode phase images (Fig. 9, right) substantially
confirmed the results obtained with the height modulated tech-
nique but allowed for a more accurate determination of the surface
parameters, which are collected in Table 5. The blend of P3HT with
T6F showed the lower root mean square (RMS) roughness and
average diameter of the grains while the blend with C60-fullerene
evidenced the higher roughness and diameter of crystallites,
probably ascribable to nanoaggregates of fullerene. The phase
image of P3HT/PCBM blend showed a few 14 nm average diameter
nanocrystallites of PCBM on an essentially amorphous surface,
indicating that the addition of PCBM can perturb the nanoscale
crystallinity usually observed for pure P3HT films [59].
4. Conclusions
In this work we synthesized a new 3-substituted thiophenic
monomer bearing a C60-fullerene group at the end of its side chain.
This monomer was copolymerized with an alkylthiophene deriva-
tive, leading to a copolymer that has high fullerene content (1:1 wt/
wt) but which is still soluble in common organic solvents. The
thermal, photophysical and morphological behaviors of the
obtained copolymer were analyzed and compared with those of
a reference polymer usually used for polymeric solar cells, indi-
cating the importance of the solvent used to obtain the polymer
films as well as the necessity to optimize the annealing procedure.
Both the monomer blended with P3HT and the copolymer were
tested as photoactive layers for the construction of BHJ solar cells
and their performances were compared with other cells prepared
with standard materials and assembled with the same architecture.
In the case of the double-cable copolymer, good control of the
nanophase separation made it possible to obtain a photovoltaic
efficiency that was higher than with the reference compounds; an
even higher value was obtained by using the fullerene-
functionalized monomer in blend with P3HT instead of the
usually employed PCBM or C60-fullerene. These results clearly
indicate that the newly synthesized monomer and copolymer can
be strategic materials for organic photovoltaic applications and
a study is currently in progress to synthesize a fully regioregular
copolymer to evaluate the effects of the configurational regularity
on the device performance.
References
[1] Peet J, Senatore ML, Heeger AJ, Bazan GC. Adv Mater 2009;21:1521e7.
[2] Helgesen M, Sondergaard R, Krebs FC. J Mater Chem 2010;20(1):36e60.
[3] Brabec CJ, Sariciftci NS, Hummelen JC. Adv Funct Mater 2001;11:15e26.
[4] Park SH, Roy A, Beaupre S, Cho S, Coates N, Moon JS, et al. Nat Photon 2009;3:
297e303.
[5] Kim JY, Kim SH, Lee HH, Lee K, Ma W, Gong X, et al. Adv Mater 2006;18:572e6.
[6] Ma W, Yang C, Gong X, Lee K, Heeger AJ. Adv Mater 2005;15:1617e22.
[7] Salaneck WR, Inganas O, Themans B, Nilsson JO, Sjogren B, Osterholm JE, et al.
J Chem Phys 1998;89:4613e9.
[8] Bao Z, Dodabalapur A, Lovinger AL. Appl Phys Lett 1996;69:4108e10.
[9] Padinger F, Rittber RS, Sariciftci NS. Adv Funct Mater 2003;13:85e8.
[10] Chen D, Nakahara A, Wie D, Nordlund D, Russell TP. Nano Lett 2011;11:
561e7.
[11] Reyes-Reyes M, Kim K, Carroll DL. Appl Phys Lett 2005;87: 083506e1.
[12] Hoppe H, Sariciftci NS. J Mater Chem 2006;16:45e61.
[13] Li MH, Li XF, Xu P, Yang SF. Macromol Chem Phys 2010;211:443e52.
[14] Mihailetchi VD, Xie H, De Boer B, Koster LJA, Blom PWM. Adv Funct Mater
2006;16:699.
Table 5
AFM surface data and photovoltaic properties of the assembled solar cells.
d
Polymer
Roughness Diameter VOC
JSC
FFc
h (%)
a
b
(nm)
(nm)
(V)
(mA ꢂ cmꢀ2
)
P3HT/C60
P3HT/PCBM 0.53
CP1
P3HT/T6F
3.26
16
14
16
13
0.41 4.31
0.48 5.00
0.50 6.20
0.60 7.20
0.41 0.72
0.48 1.08
0.51 1.55
0.52 2.24
1.92
0.40
a
Open circuit voltage.
Short-circuit current density.
Fill factor.
[15] Kim H, So WW, Moon SJ. Sol Energ Mater Sol Cells 2007;91:581e7.
[16] Cravino A. Polym Int 2007;56:943e56.
[17] Jeffries M, Sauvé G, McCullough RD. Adv Mater 2004;16:1017e9.
[18] Bauerle P, Wurthner F, Heid S. Angew Chem Int Ed Engl 1990;29:419e20.
b
c
d
Power conversion efficiency.

M. Lanzi et al. / Polymer 53 (2012) 2134e2145
2145
[19] Mohamad DK, Chauhan SS, Yi H, Cadby AJ, Lidzey DG, Iraqi A. Sol Energ Mater
Sol Cells 2011;95:1723e30.
[20] Cravino A, Zerza G, Maggini M, Bucella S, Svensson M, Andersson MR, et al.
Monatshefte für Chemie 2003;134:519e27.
[21] Yassar A, Hmyene M, Loveday DC, Ferraris JP. Synth Met 1997;84:231.
[22] Prato M, Maggini M. Acc Chem Res 1998;31:519e26.
[23] Costa Bizzarri P, Andreani F, Della Casa C, Lanzi M, Salatelli E. Synth Met 1995;
75:141e7.
[24] Costa Bizzarri P, Della Casa C, Lanzi M, Bertinelli F, Iarossi D, Mucci A, et al.
Synth Met 1999;104:1e7.
[41] Robitaille L, Leclerc M, Callender CL. Chem Mater 1993;5:1755e61.
[42] Kesters E, de Kok MM, Carleer RAA, Czech JHPB, Adriaensens PJ, Gelan JM,
et al. Polymer 2002;43:5749e55.
[43] Ouhib F, Khoukh A, Ledenil JB, Martinez H, Desbrières J, Dagron-Lartigan C.
Macromolecules 2008;41:9736e43.
[44] Bondouris BW, Molins F, Blank DA, Frisbie CD, Hillmyer MA. Macromolecules
2009;42:4118e26.
[45] Cugola R, Giovanella U, Di Gianvincenzo P, Bertini F, Catellani M, Luzzati S.
Thin Solid Films 2006;511e512:489e93.
[46] Chirvase D, Parisi J, Hummelen JC, Dyakonov VV. Nanotechnology 2004;15:
1317.
[47] Pakhomov L, Drozdov Yu N. Cryst Eng 2004;6:23.
[48] Swinnen A, Haeldermans I, Van de Ven M, D’Haen J, Vanroyland G, Aresu S,
et al. Adv Funct Mater 2006;16:760e5.
[25] Andreani F, Salatelli E, Lanzi M. Polymer 1996;37(4):661e5.
[26] Cravino A, Serdar Sariciftci N. J Mater Chem 2002;12:1931e43.
[27] Zhang F, Svensson M, Andersson MR, Maggini M, Bucella S, Menna E, et al. Adv
Mater 2001;13(24):1871e4.
[28] Hong Xiaoyong, Tyson Cameron J, Middlecoff JS, Collard DM. Macromolecules
1999;32:4232e9.
[49] Ma WL, Yang CJ, Gong X, Lee K, Heeger AJ. Adv Funct Mater 2005;15:
1617e22.
[29] Singh RK, Kumar J, Singh R, Kant R, Chand S, Kumar V. Mater Chem Phys 2007;
104:390e6.
[50] McCullough RD, Tristam-Nagle S, Williams SP, Lowe RD, Jayaraman M. J Am
Chem Soc 1993;115:4910e1.
[30] Wei H, Scudiero L, Eilers H. Appl Surf Sci 2009;255:8593e7.
[31] Hotta S, Rughooputh SDDV, Heeger AJ, Wudl F. Macromolecules 1987;20:212e5.
[32] Vilar Rei M, Schott M, Pireaux JJ, Grégoire C, Thiry PA, Caudano R, et al. J Surf
Sci 1987;189-90:927e34.
[33] Louarn G, Trznadel M, Buisson JP, Laska J, Pron A, Lapkowski M, et al. J Phys
Chem 1996;100:12532e9.
[34] Zerbi G, Chierichetti B, Inganas O. J Chem Phys 1991;94:4646e58.
[35] Li M, Xu P, Yang J, Yang S. J Mater Chem 2010;20:3953e60.
[36] Della Casa C, Costa Bizzarri P, Lanzi M, Paganin L, Bertinelli F, Pizzoferrato R,
et al. Synth Met 2003;138:409e17.
[37] Zagorska M, Kulszewicz-Bajer I, Pron A, Sukiennik J, Raimond P, Kajzar F, et al.
Macromolecules 1998;31:9146e53.
[51] Chen T, Wu X, Rieke RD. J Am Chem Soc 1995;117:233e44.
[52] Kim DH, Park YD, Jang Y, Yang H, Cho K, Kim YH, et al. Adv Funct Mater 2005;
15:77.
[53] Jordan EF, Feldeisen DW, Wrigley AN. J Polym Sci Part 1 1971;9:1837.
[54] Yamamoto T. Bull Chem Soc Jpn 1999;72:621.
[55] Heyding RD, Russel KE, Larty TL. Powder Diffr 1990;5:93.
[56] Iovu MC, Sheina EE, Gil RR, McCullough RD. Macromolecules 2005;38:
8649e56.
[57] Wu X, Chen TA, Rieke RD. J Am Chem Soc 1995;117:233.
[58] Yamamoto T, Kumaruden D, Arai M, Lee BL, Suganuma H, Asakawa N, et al.
J Am Chem Soc 1998;120:2047.
[59] Tremolet de Villers B, Tassone CJ, Tolbert SH, Schwartz BJ. J Phys Chem C 2009;
113:18978e82.
[38] Haba O, Hayakawa T, Ueda M, Kawaguchi H, Kawazoe T. React Funct Polym
1998;37:163e8.
[39] Tanatar MA, Graja A, Zhu DB, Li YL. Synth Met 1998;94:83e6.
[40] Tan Z, Hou Y, He Y, Zhou E, Yang C, Li Y. Macromolecules 2007;40:1868e73.
[60] Brabec CJ, Padinger F, Dyakonov V, Hummelen JC, Janssen RAJ, Sariciftci NS. In:
Electronic properties of novel materials-Progress in molecular nanostructures.
Singapore: World Scientific Co; 1998. p. 519 [ff].