Use of poly(3-methylthio)thiophene blends for direct laser tracing and bulk heterojunction solar cells

Article abstract of DOI:10.1016/j.reactfunctpolym.2014.07.007

In this article we demonstrate the use of a blend made of two regioregular polythiophenic derivatives, namely poly(3-methylthio)thiophene and poly(3-hexyl)thiophene, to obtain conductive traces by the simple laser exposure of their thin films to a suitable laser source. The polymeric blend was also tested as a photoactive layer for BHJ solar cells, showing an improved surface morphology and a wider absorption spectrum, thus resulting in an enhanced photovoltaic performance. In the standard condition normally used for the cell preparation, we obtained a 3.16% power conversion efficiency. The device showed good reproducibility and stability over time.

Full text of DOI:10.1016/j.reactfunctpolym.2014.07.007

Reactive & Functional Polymers 83 (2014) 33–41
Contents lists available at ScienceDirect
Reactive & Functional Polymers
journal homepage: www.elsevier.com/locate/react
Use of poly(3-methylthio)thiophene blends for direct laser tracing
and bulk heterojunction solar cells
a
b
a
a
Massimiliano Lanzi a, , Luisa Paganin , Filippo Pierini , Francesco Errani , Francesco Paolo Di-Nicola
⇑
a Dipartimento di Chimica Industriale ‘‘Toso Montanari’’, Università di Bologna, Viale del Risorgimento, 4 I-40136 Bologna, Italy
b Dipartimento di Chimica ‘‘G. Ciamician’’, Università di Bologna, Via Selmi, 2 I-40126 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 article we demonstrate the use of a blend made of two regioregular polythiophenic derivatives,
namely poly(3-methylthio)thiophene and poly(3-hexyl)thiophene, to obtain conductive traces by the
simple laser exposure of their thin films to a suitable laser source. The polymeric blend was also tested
as a photoactive layer for BHJ solar cells, showing an improved surface morphology and a wider absorp-
tion spectrum, thus resulting in an enhanced photovoltaic performance. In the standard condition nor-
mally used for the cell preparation, we obtained a 3.16% power conversion efficiency. The device
showed good reproducibility and stability over time.
Received 5 August 2013
Received in revised form 6 June 2014
Accepted 5 July 2014
Available online 17 July 2014
Keywords:
Electrical conductivity
Laser tracing
Ó 2014 Elsevier Ltd. All rights reserved.
Bulk heterojunction polymeric solar cells
Regioregular polyalkylthiophenes
Polymer blends
1. Introduction
[3], inkjet printing [4] and roll-to-roll methods [5], making them
very interesting materials from an industrial standpoint.
The need for flexible, lightweight, easily processable new mate-
rials for electronics has pushed research toward the synthesis of
organic materials able to compete with the ‘‘classic’’ inorganic
semiconductors in their own application fields. Organic semicon-
ductors, such as conjugated polymers, are promising candidates
for this replacement, as demonstrated both by the number of pat-
ent applications in 2013 (more than 100 on polythiophenes) and
by the growing number of newly launched companies and existing
manufacturers of materials or devices that have added organic
photovoltaics to their portfolio.
In this context, polythiophenes are intriguing materials which
have now been studied for a long time thanks to their electric
and electronic properties. In fact, in the charged (doped) state they
are very effective conductors of electricity, while in the neutral
(undoped) state they are mainly employed in optoelectronic
devices – for the construction of organic light emitting diodes
(OLEDs), low-voltage field effect transistors and bulk heterojunc-
tion (BHJ) solar cells – and in devices such as optical filters, signal
modulators and polarization rotators. Alkyl-substituted polythio-
phenes (PATs) are very soluble in common organic solvents and
can be easily filmed and processed using a number of different
techniques, e.g. spincoating [1], doctor blading [2], screen printing
Structurally, PATs belong to the class of conjugated polymers
the precursor of which is polyacetylene (PAc). PAc shows electronic
characteristics similar to those of PATs; recently some researchers
have reported that PAcs functionalized with a thioalkylic group
(PAc-SR) are photosensitive polymers [6]. In fact, thin insulating
films of PAc-SR can be traced by laser operating at suitable speed,
power, and wavelength leading to electroconductive patterns on
the film surface [7]. This approach is particularly intriguing since
it makes it technically feasible to obtain high resolution patterns
for electronic circuits while strongly limiting the number of pro-
cessing steps and chemicals required. In fact, the line patterning
of an insulating polymeric matrix by a laser-induced photopyroly-
sis process makes it possible to rapidly develop custom conductive
circuits from a schematic drawing, by using simple CAD (computer
aided design) systems to drive the laser source [8]. Various advan-
tages are evident: over conventional inorganic conductors (metals,
oxides, silicon), since no multiple etching and lithographic steps
are required to obtain the final circuit, and over the most synthe-
sized conjugated polymers, as no redox processes are necessary
to make the polymer conductive, thus avoiding the well-known
problems of time and environmental stability.
But conjugated polymers are also intensively studied for other
technological applications. In fact, in recent years, conjugated poly-
mers-fullerene mixtures have been widely investigated for use in
organic solar cells. Bulk heterojunction (BHJ) polymer solar cells
⇑
Corresponding author. Tel.: +39 051 2093689; fax: +39 051 2093669.
E-mail address: massimiliano.lanzi@unibo.it (M. Lanzi).
http://dx.doi.org/10.1016/j.reactfunctpolym.2014.07.007
1381-5148/Ó 2014 Elsevier Ltd. All rights reserved.

34
M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
have been reported using either poly(p-phenylenevinylene)
derivatives [9], poly(3-alkylthiophene)s [10,11], or low-bandgap
polymers [12,13] as electron-donors combined with PCBM ([6,6]-
phenyl-C61-buthyric acid methyl ester), a soluble fullerene deriva-
tive, as the electron acceptor thus creating the photoactive blend.
In the standard architecture, the photoactive blend is sandwiched
between a glass covered with indium tin oxide (ITO), acting as the
anode, and a cathode made of a low-workfunction metal (usually
Al). The light incident on the BHJ cell through the ITO electrode is
mainly absorbed by the conjugated polymer, thus leading to the cre-
ation of bound excitons (holes/electrons couples). If the polymer
and fullerene components are phase-segregated on a nanoscale
length, the excitons can split into separated electrons and positive
holes, which are transported along PCBM and conjugated polymer
toward the metal and ITO electrode, respectively, thus generating
the photocurrent and photovoltage.
One factor that strongly limits the efficiency of this kind of cell
is the low exploitation of the sunlight due to the narrower absorp-
tion band of the absorption spectrum of the conjugated polymer in
comparison to the solar spectrum; however some strategies have
been adopted to overcome this limitation, in particular the synthe-
sis of low band-gap polymers [14–16] and the preparation of poly-
thiophenes containing chromophores in the side chain [17,18].
Therefore, the aim of this work is to synthesize a regioregular
polythiophenic derivative functionalized with thiomethylic groups
– namely the poly(3-methylthio)thiophene (PTSMe) – and to study
its electrical behavior before and after laser tracing. For this pur-
pose, PTSMe was blended with regioregular poly(3-hexyl)thio-
phene (P3HT) to overcome its poor filmability caused by its
incomplete solubility in organic solvents. The obtained blend
showed a wider absorption spectrum than conventional P3HT
and was employed for the preparation of BHJ solar cells. The
assembled cells were fully characterized from the morphological
and electronic standpoint and their performances was compared
with a reference cell made of only P3HT.
stirring and in an inert atmosphere – to a solution of 2.0 g
(15.4 mmol) of 3-methylthiothiophene (TSMe, Atlantic Chemical
Co., USA) in 15.5 ml of DMF. The mixture was reacted for 6 h at
room temperature. 3.02 g (17.05 mmol) of NBS in 17.1 ml of DMF
were added dropwise and the mixture was reacted for another
24 h at room temperature and under nitrogen. The reaction mix-
ture was then poured into 500 ml of an aqueous solution of NaCl
and the organic phase was extracted with 4 Â 150 ml of n-pentane.
The collected organic phases were washed with a 5% solution of
NaHCO3 (2 Â 300 ml) and with water to neutrality. The organic
phase was then dried and concentrated at reduced pressure thus
giving the crude product, which was purified with column chroma-
tography (SiO2/n-heptane) leading to 3.19 g of pure 2,5-BTSMe
(72% yield).
1H NMR (CDCl3, ppm) d: 6.91 (s, 1H, H4); 2.55 (s, 3H, CH3).
FT-IR (Ge disk, cmÀ1): 3051 (
m
CAH, thiophene b-hydrogen);
CAH, antisymmetric methyl); 2875 ( CAH, symmetric
methyl); 1503 ( C@C, antisymmetric thiophene); 1468 ( C@C,
symmetric thiophene); 1368 (symmetric deformation CH3); 991
CABr, thiophenic); 840 ( CAH 2,3,5-trisubstituted thiophene);
690 ( CASMe).
2960 (
m
m
m
m
(m
c
m
2.1.2. Synthesis of the regioregular poly(3-methylthio)thiophene
(PTSMe)
4.2 ml (4.2 mmol) of a 1 M solution of methylmagnesium bro-
mide in di-n-buthyl ether were added to a solution of 1.20 g
(4.16 mmol) of 2,5-BTSMe in 25 ml of anhydrous THF. The mixture
was refluxed for 2 h under stirring and under Ar atmosphere and
then 24.83 mg (0.0458 mmol) of Ni(dppp)Cl2 were added and the
reaction refluxed for 1 h more. The mixture was cooled down to
room temperature, poured into 40 ml of methanol and filtered on
a Teflon septum (0.45 lm pore size). The recovered polymer was
washed several times with methanol and dried, resulting in
0.320 g of PTSMe (60 % yield).
1H NMR (CDCl3, ppm) d: 7.26 (s, 1H, H4); 2.49 (s, 3H, CH3).
FT-IR (Ge disk, cmÀ1): 3059 (
m
CAH, thiophene b-hydrogen);
CAH, antisymmetric methyl); 2853 ( CAH, symmetric
methyl); 1511 ( C@C, antisymmetric thiophene); 1462 ( C@C,
symmetric thiophene); 1369 (symmetric deformation CH3); 1014
(rocking CH3); 823 ( CAH 2,3,5-trisubstituted thiophene); 716
CASMe).
2957 (
m
m
2. Experimental
m
m
2.1. Synthesis and polymerization
c
(m
Scheme 1 outlines the experimental route from monomer to
polymer.
2.1.3. Preparation of the PTSMe-P3HT blend (PB)
120 mg (0.94 mmol) of PTSMe and 60 mg (0.36 mmol) of regio-
regular poly(3-hexylthiophene) (P3HT, 92% HT, Mn = 32 kDa,
PDI = 1.18), previously synthesized using the same procedure used
for the preparation of PTSMe, were dissolved in 10 ml of CHCl3 at
room temperature and the obtained solution was sonicated for
10 min. The solution was then filtered over a Teflon septum
2.1.1. Synthesis of the monomer 2,5-dibromo-3-methylthiothiophene
(2,5-BTSMe)
2.73 g (15.4 mmol) of N-bromosuccinimide (NBS) in 15.5 ml of
N,N-dimethylformamide (DMF) were added dropwise - under
(0.20 lm pore size) and concentrated at reduced pressure produc-
ing the PB blend.
SCH3
Br
SCH3
NBS, DMF
2.2. Measurements
Br
S
S
1H NMR were recorded on a Varian Mercury Plus spectrometer
(400 MHz) using TMS as a reference. FT-IR spectra of the mono-
mers (pure liquids) and polymers (films) were obtained on Ge
disks using a Perkin Elmer Spectrum One spectrophotometer.
Molecular weights were determined by gel permeation chroma-
tography (GPC) using polystyrene standards and THF as an eluent
on a HPLC Lab Flow 2000 apparatus equipped with a PL Gel MXL
column and a Linear Instrument UV–Vis detector model UVIS-
200 working at 263 nm. UV–Vis spectra were recorded on a Perkin
Elmer Lambda 19 spectrophotometer using polymer films on
quartz slides cast from chloroform solutions (ca. 10À3 M). Cyclic
TSMe
2,5-BTSMe
SCH3
CH3MgBr, Ni(dppp)Cl2
THF
S
n
PTSMe
Scheme 1. Synthesis of the monomer 2,5-BTSMe and polymer PTSMe.
voltammetry (CV) was performed on a Autolab PGSTAT 30

M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
35
(EcoChemie, The Netherlands) with a three electrode system in a
were quite difficult to find but, when the system was correctly
set-up, conducting traces became insoluble in common solvents
and insensitive to humidity, chemical vapors and high temperature.
When exposed to a coherent light source, these polymers undergo a
photopyrolysis process, with the partial loss of the thioalkylic sub-
stituent, which leads to the formation of unsaturated backbones
mainly made up of sp2 hybridized carbon atoms which are only
partly linked with sulfur substituents in side chains [19]. This pro-
solution of 0.1 M Bu4NPF6 in CH3CN, at a scan rate of 50 mV sÀ1
.
Polymer films were coated on a Pt plate electrode (1 cm2) by dip-
ping the electrode into the corresponding solution (5 Â 10À3 M in
CHCl3) and then drying, obtaining films of the approximatively
same thickness (0.4–0.5 lm). A Pt wire was used as the counter
electrode and an aqueous saturated calomel electrode (SCE) as
the reference electrode.
Electrical measurements were performed in air at room temper-
ature using a Keithley 2101 electrometer (traced films) and an Alpha
Lab teraohmeter (pristine films). The reported values were the mean
of five measurements performed on different parts of the same sam-
ple as well as on five different samples prepared using the same
experimental conditions. In all cases, differences did not exceed
5% of the final value. The laser source was a Wicked Lasers Spyder
III Arctic class 4 diode laser, operating at 445 nm with a nominal
power of 750 mW. The effective power of the laser beam on the
polymer surface was measured with a Coherent FieldMax II laser
power meter giving an effective power of 600 mW. Samples were
mounted on a computer-controlled positioning system (Thorlabs
L490MZ) and moved on a plane perpendicular to the focused laser
beam by using two Thorlabs MTS TDC001 controllers thus making
it possible to control both the tracing speed on the x–y plane and
the laser-sample distance for the correct focus of the beam on the
sample surface, while operating on the z-axis. Films thickness was
measured using an AFM Burleigh Vista 100 as a profilometer. SEM
analyses were performed on a Phenom-world ProX SEM apparatus,
equipped with an EDS microanalysis probe. AFM of the blend was
performed on a Burleigh Vista Atom Force Microscope equipped
with a silicon–nitride tip and operating in a non-contact tapping
mode. Optical 2D and 3D microscopy was performed using a Hirox
KH-7700 Digital microscope.
cess leads to an extended p-system delocalized over the polymeric
chain with an enhanced main chain mean conjugation length
(MCL), which explains the higher electrical conductivity.
In this work, we tried to substitute PAc-SMe [20] with an easier
synthesizable polythiophenic derivative, i.e. poly(3-methylthio)thi-
ophene (PTSMe), which possesses a photosensitive functional group
and a polyconjugated backbone, similar to the acetylene-based
polymer, but also shows an increased solubility, filmability and
environmental resistance in the non-traced pristine state. Since reg-
ioregularity is a fundamental prerequisite for obtaining PATs with
high electron mobility [21], PTSMe was prepared using the McCul-
lough Grignard Metathesis (GRIM) polymerization procedure
(Scheme 1), which generally leads to regioregular and soluble thio-
phenic polymers [22].
This procedure starts from a 2,5-dihalothiophene derivative and
leads to the obtainment of functionalized polythiophenes with a
good yield and a high degree of regioregularity through a simple
and effective organometallic group exchange (Grignard Metathesis)
reaction. With this aim, the monomer 2,5-dibromo-3-(methyl-
thio)thiophene (2,5-BTSMe) was prepared through the dibromina-
tion of TSMe with N-bromosuccinimide (NBS) in anhydrous N,N-
dimethylformamide (DMF). The halogenation conditions were
accurately optimized, providing for the addition of NBS in two sep-
arate amounts while operating at room temperature, thus leading to
a satisfying yield (72%) of 2,5-BTSMe. We tried also a faster way,
which consisted of the one-step addition of the entire amount of
NBS and a reaction time of only 4 h at 60 °C. This time, however,
we observed a partial bromination of the ACH3 in the side chain.
2.2.1. ITO/PEDOT/PB: PCBM/Al solar cell assembly
ITO glass (Delta Technologies, Stillwater, Minnesota, USA;
2.5 Â 2.5 cm; code CG-41IN-0107) was first cleaned in an ultrasonic
bath using a non-foaming glass detergent in deionized water. ITO
glass was then rinsed sequentially in double distilled water, isopro-
The two bromine atoms in the
a-positions of 2,5-BTSMe were
exploited for the organometallic coupling involving the magne-
sium-halogen exchange (metathesis reaction) with a preformed
Grignard derivative (1st reaction step), followed by a Ni(II) cata-
lyzed cross-coupling reaction (2nd reaction step), thus leading
directly to the regioregular polymer. These two steps are consecu-
tive and the polymerization is a simple and fast one-pot process.
PSMe was then obtained as a red-brownish powder highly sol-
uble in aromatic chlorinated solvents (chlorobenzene and o-
dichlorobenzene) up to 30 mg/ml, but its solubility decreases to
about 5–10 mg/ml in common organic solvents (CHCl3, THF) at
room temperature. This can be ascribed to PTSMe short side chains,
which are not able to produce a strong plastifying effect, and not to
the polymer main chain lengths, since its molecular weight was
not particularly high (Mn = 18.000, PDI = 1.2).
Films of PTSMe cast on glass slides or PET foils show poor
homogeneity, some dots, and brittleness and do not have good
and permanent adhesion on different surfaces. In order to over-
come this problem, we blended the prepared PTSMe with a previ-
ously synthesized P3HT sample prepared with the same
polymerization technique (GRIM procedure), since P3HT is well
soluble in common organic solvents, from which it produces thick,
homogeneous free-standing films, while its optical and electrical
properties have been well known for a long time now.
panol, and acetone reaching a final resistance of 6
X/sq. PEDOT:PSS
(Aldrich Chemical Co.) was diluted 1:1 with isopropanol and depos-
ited by doctor blading (DB) on top of the cleaned ITO glass using a
Sheen Instruments Model No. S265674 (film thickness about
80 nm). Anhydrous chloroform was used to prepare solutions of
P3HT or PB and PCBM (1:1 weight ratio), which were deposited
by DB on the PEDOT:PSS layer. After baking films in a vacuum at
130 °C for 30 min, the active layer film thickness measured by
AFM was about 100 nm. Lastly, to create the OPV devices, 50 nm
of Al was thermally deposited under a vacuum of 6 Â 10À7 mmHg
by means of an Edwards E306A vacuum coating apparatus
equipped with a dual-stage mechanical vacuum pump Edwards 2
E2H2 and with a selectable diffusion pump or turbo-molecular
pump. The current–voltage characteristics were measured using a
Keithley 2401 source meter under the illumination of a Abet
Technologies LS 150 Xenon Arc Lamp Source AM1.5 Solar Simulator,
calibrated with an ILT 1400-BL photometer.
3. Results and discussion
Polyacetylenes functionalized with thiomethylic (PAc-SMe) or
thioethylic (PAc-SEt) groups were the first photosensitive polymers
examined for laser patterning applications. In fact, they were able
to increase their electrical conductivity by many orders of magni-
tude (up to 16) when exposed to a correctly focused laser radiation
of a suitable wavelength (351 or 488 nm) without need for a chem-
ical doping. The best conditions for the tracing of PAc derivatives
The polymeric blend (PB) was prepared starting from a 2:1
weight ratio (about 7:3 M ratio) between PTSMe and P3HT. The
FT-IR analysis of PB was performed on a thin film cast on a Ge disk
from CHCl3 solution (Fig. 1).
The spectrum clearly shows the peaks ascribable to the
presence of both polymers. Most of them are common to both

36
M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
the thiomethylic protons, while the triplet at 2.80 ppm (ACH2
a
to the thiophene ring), the multiplet at 1.73 ppm (ACH2 b to the thi-
ophene ring), the broad multiplet in the 1.50–1.20 ppm range
(ACH2 of the central units) and, lastly, the triplet at 0.90 ppm
(ACH3) are due to the protons of the P3HT side chains. The compo-
sition of the blend can be determined by means of the normalized
integral ratio of the signals at 2.49 and 0.90 ppm and, alternatively,
by the normalized integral ratio of the two signals at 7.26 and
6.99 ppm. The real composition of PB was quite different from the
starting ratio of the two polymers. In fact, the final molar ratio
between PTSMe and P3HT was around 1:2 instead of the 7:3 feed
ratio used. This can be ascribed to the low solubility of PTSMe in
common organic solvents which caused a kind of fractionation
when we filtered the PB solution on the Teflon septum.
However, PB was well soluble in CHCl3, producing homoge-
neous and glossy dark-red films.
Fig. 3 shows the UV–Vis spectra of the polymers in film cast
from CHCl3 solutions. Regioregular P3HT shows a kmax at 554 nm,
corresponding to the first vibronic quantum, and an evident shoul-
der at 595 nm, corresponding to the pure electronic transition E0-0
[23] which derives from the vibronic absorption of ordered P3HT
crystalline regions in films [24]. PTSMe shows only one peak at
415 nm, blue-shifted with respect to the kmax of P3HT and does
not show any evident vibronic transition, since short side chains
are not able to interact with themselves, thus giving more ordered
main chains conformations [25]. The spectrum of the blend shows
a very large absorption region, starting at 350 nm and ending at
about 650 nm. The blend spectrum clearly evidences the main
absorptions of its two components: the PTSMe absorption maxi-
mum is now embedded in the spectrum and becomes visible as a
shoulder at 410 nm, while the transitions ascribable to the P3HT
component are found at 599 nm (E0-0), 546 nm (first vibronic
quantum), and 488 nm (kmax). The broad absorption range of the
blend can be very useful for polymeric solar cells, since a wider
light absorption in the visible region of the solar spectrum can gen-
erate more excitons in the BHJ active layer, thus increasing the Jsc
of the final device [26].
Fig. 1. FT-IR spectrum of PB in film cast on Ge disk from CHCl3 solution. The C-SMe
stretching band is evidenced by an asterisk.
the polymers but some peaks can be ascribed to each single struc-
ture. In detail, the signals (in cmÀ1) at 3055 (
m
CAH, thiophene b-
CAH, antisymmetric methyl), 2856 ( CAH,
symmetric methyl), 1510 C@C, antisymmetric thiophene),
1455 ( C@C, symmetric thiophene), 1377 (symmetric deformation
CH3) and 820 ( CAH 2,3,5-trisubstituted thiophene) are common
to both the polymers while the signals at 2925 ( CAH, antisym-
metric methylenes) and 757 (CH2 rocking) are characteristics of
hydrogen), 2955 (
m
m
(
m
m
c
m
P3HT and that at 716 (m CASMe) is exclusive to PTSMe.
1H NMR spectrum of PB in CDCl3 is reported in Fig. 2.
Starting from the aromatic region, the peak at 7.26 ppm is
ascribable to the thiophenic H-4 proton of the TSMe monomer
and the fact that it is a singlet confirms that we obtained a regioreg-
ular PTSMe polymer. The signal at 6.99 ppm is ascribable to the aro-
matic H-4 protons of the T6H moieties; this time also, the spectrum
confirms the high regioregularity of the P3HT sample, since the
intensity of the peak attributable to the HT-HT dyads is strongly
prevalent over those belonging to the other kinds of linkages, also
visible at about 7.00 ppm. The singlet at 2.49 ppm is ascribable to
A few drops of PB solution (5 mg in 5 ml) in CHCl3 were depos-
ited on a quartz slide (4.0 Â 4.0 cm) using the DB technique thus
Fig. 2. 1H-NMR spectrum of PB in CDCl3. CHCl3 peak is indicated by an asterisk.

M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
37
Fig. 5. 3D image of PB traces on the polymer film cast from CHCl3 solution.
Fig. 3. UV–Vis spectra of polymers in film cast from CHCl3 solutions.
Fig. 6. SEM micrograph of a PB film after laser tracing. The polymer film was cast
from PB solution in CHCl3.
Fig. 4. Optical microscope image of a traced PB sample (Â50). The film was cast
from the polymer solution in CHCl3.
giving an homogeneous 1.5 lm thick film that was used for the
laser tracing procedure. The best tracing conditions were found
at a 1.0 mm/s tracing speed, by making three contiguous parallel
lines with an inter-trace distance of 0.33 mm (Fig. 4).
The obtained traces are clearly visible since the polymer surface
changed notably after the laser exposure, going from a very
smooth, homogeneous surface to a rough, porous surface. This is
also evident in the 3D image and in the SEM micrograph of Figs. 5
and 6, respectively.
Film porosity is determined by the outflow of gaseous reaction
products – mainly sulfides and mercaptanes [27] – which leaves
the polymer surface during the laser patterning.
Fig. 7. FT-IR spectrum (expansion) of a PB film on Ge disk from CHCl3 solution
before and after the laser tracing procedure. The C-SMe stretching band is
evidenced by an asterisk.
The band at 716 cmÀ1 in the IR spectrum of PB, ascribable to
CAS stretching [28], decreases in intensity after the laser tracing,
thus confirming the partial loss of the thiomethylic group after
the photopyrolysis procedure (Fig. 7).
We observed that the resistance of the sample decreases as the
number of contiguous traces increases, rapidly reaching an asymp-
totic value corresponding to about three lines. A droplet of an Ag
conductive paint (Ted Pella Inc., USA) was then applied to the
two ends of the 3-cm-long trace and the resistance was measured
with a Keithley 2401 source meter. We prepared five samples of
the traced PB blend and measured their surface resistance, which
The surface resistance of the samples before tracing was
210 GX/sq ( 5%) and was measured with a teraohmeter (Alpha
Lab Inc., USA) using a two probe measuring system. Samples and
probes were carefully shielded from surrounding EMF with a Fara-
day-cage homemade device.
We then prepared another series of samples under the same
experimental conditions, i.e. by depositing the CHCl3 solution of
PB on a quartz slide using DB technique. This time, samples were
annealed for 2 h at 130 °C under a vacuum (0.5 mmHg) by using
was 150 kX/sq with a very good reproducibility of this value
among the samples, with a deviation from the reported value of 3%.

38
M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
thiomethylic side chains. It is worth noting that this slight changes
in the PB composition due to the laser exposure is capable of deter-
mining such a high surface resistivity change.
Lastly, we prepared a new set of samples made of P3HT alone, in
order to evaluate whether the laser treatment was capable of
decreasing their electrical resistance. P3HT films were deposited
by DB using a polymer solution in CHCl3. In this case, the surface
resistance of these samples was the same before and after the
exposure to laser radiation i.e. 220 GX/sq ( 6%). This test confirms
that the electrical conductivity enhancement is ascribable to the
presence of the photo-sensitive thiomethylic groups, which are
involved in a laser-induced photocleavage which leaves the poly-
thiophenic backbone able to partially crosslink and, then, to rear-
range in a more planar and more conjugated conformation, thus
aiding the electroconductivity of the final material [29].
Fig. 10 shows the J–V characteristic curves of solar cells which
have the structure of ITO/PEDOT:PSS/composite (photoactive poly-
mer:PCBM)/Al. PEDOT:PSS acts as a hole-transporting layer, the
photoactive polymer (P3HT or PB) as an electron donor, and PCBM
as an electron acceptor, while the Al electrode is the cathode of the
solar cell and the ITO layer the anode [30]. The operational proce-
dures followed for the preparation of cells are described in the
experimental section. We used a polymer/PCBM 1:1 weight ratio
since some studies have shown that devices with this ratio achieve
the highest power conversion efficiency [31,32] and the film thick-
nesses of the blend absorber layers were each about 100 nm, in
order to make the measurement of the photocurrent independent
of the thickness of the samples.
Fig. 8. SEM micrograph of a pre-annealed PB film, cast from CHCl3 solution, after
laser tracing.
a Büchi B-585 glass oven before the laser tracing. Even if their ini-
tial resistance was essentially unchanged (about 200 G
the final resistance of the traced samples was 27 2% k
X
/sq ( 4%)),
/sq. This
X
time, the laser treatment of samples enabled a resistance decrease
of six orders of magnitude. It must be stressed that the annealing
procedure was also performed on the previously traced samples
but – this time – it was ineffective, since no evident changes in
the final resistance values were found. In fact, the exposure to
the laser light makes the traced polymer apparently harder, more
scratch-resistant, and insoluble in organic solvents, thus suggest-
ing that the exposed polymer no longer has the characteristics of
The device parameters: short circuit current (Jsc), open circuit
voltage (Voc), Fill Factor (FF), and power conversion efficiency
(PCE) are summarized in Table 1.
The electrochemical behavior of the three examined polymers
has been evaluated performing cyclic voltammetry (CV) on their
thin films cast on a Pt electrode from CHCl3 solutions (Fig. 11).
We decided to characterize also a film of PSMe even if, as already
mentioned, this polymer was not employed as active layer for solar
cells since its adhesion to the ITO-glass coated with PEDOT-PSS
was very poor and prevented the homogeneous deposition of the
Al cathode.
a substituted polythiophene but rather those of a system of p-con-
jugated C atoms. Fig. 8 shows the SEM micrograph of an annealed
PB sample, prepared by DB technique using PB chloroform solu-
tion, after laser tracing. Its appearance is nearly identical to the
unannealed film sample, because the laser exposure converted
the polymer into worm-like dull traces, devoid of a fine texture,
which are always apparently identical one to another.
All voltammograms exhibited reversible oxidation and a quite
good symmetry for the p-doping/undoping process, and this can
be useful for practical applications based on the charge/discharge
process since the required potential range needed for passing
between neutral and doped states is limited. The HOMO levels of
the polymers were measured by CV, while the LUMO energies were
calculated indirectly, taking into account that they correspond to
the HOMO energies plus the optical energy gaps (Eopt) evaluated
from the onsets of the UV–Vis spectra of the polymers in film
Fig. 9 shows the topographic microanalysis of the sample
shown in Fig. 8 in relation to the C and S elements recorded using
an energy-dispersive spectrometer (EDS) probe.
In Fig. 9 it is evident that the laser-exposed polymer has a
higher amount of C and a lower amount of S in the traces, since
the concentration of the examined element is proportional to the
intensity of white. More deeply, C and S concentration passed from
65% and 28% of the unexposed area to 68% and 25% of the laser-
treated portions, respectively, thus confirming the partial loss of
Fig. 9. EDS topographic microanalysis of PB film reported in Fig. 8. Left: Carbon; Right: Sulfur.

M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
39
Fig. 10. J–V characteristic curves for the prepared solar cells under AM 1.5
irradiation with an intensity of
simulator.
1 a calibrated solar
sun (100 mW/cm2) from
Table 1
Photovoltaic parameters for the devices obtained using the two different photoactive
blends.
Polymer
Jsc (mA/cm2)
Voc (V)
FF (%)
PCE (%)
P3HT
PB
7.73
8.58
0.61
0.65
52.8
56.7
2.48
3.16
[33]. The obtained values are reported in Table 2. PSMe showed the
lower oxidation potential, according to the electron-donor charac-
ter of its thiomethylic side chain directly linked to the thiophene
ring [34], and the higher bandgap, probably determined by disor-
der phenomenon like steric effects, cross linking or chain termina-
tions [35].
In Fig. 12 is reported the band diagram with HOMO/LUMO lev-
els of the ED polymers used for the solar cells in relation with the
levels of PCBM and the work function of ITO and Al, according to
Ref. [36].
PB sample showed an energy difference between its LUMO level
and the LUMO of the EA molecule (
vs 0.48 eV) and this can be favorable for the photoinduced charge
separation. In fact, it has been reported that a ELUMO = 0.3 eV is
DELUMO) higher than P3HT (0.69
D
the minimum value required to overcome the binding energy of
the excitons generated in organic semiconductor polymeric films
[37,38]. The total photovoltage of the organic solar cells is ideally
equal to the band gap value of the conjugated polymer employed
in the light-absorbing (active) layer. In the examined case, the the-
orical values should be the same for P3HT and PB since they have
the same bandgap (1.89 eV). However, there exists a loss factor
related to the high exciton binding energy in organic materials
which determines the need for a LUMO–LUMO Donor/Acceptor off-
set in D/A cells to polarize the exciton [39,40]. Practically, the
LUMO offset should be large enough to obtain an effective and fast
charge transfer and the loss factor is inversely proportional to the
Fig. 11. Cyclic voltammograms of the polymer films on Pt recorded in 0.1 M
Bu4NPF6 acetonitrile solution at a scan rate of 50 mV sÀ1
.
Table 2
Optical and electrochemical properties of the films cast from CHCl3 solutions.
onset
ox
D
ELUMO value. This could explain the Voc value of PB slightly higher
abs
edge
Polymer
Eg;opt (eV)
HOMO (eV)
LUMO (eV)
E
(V)
k
(nm)
than that of the reference cell.
PSMe
P3HT
PB
520
655
655
2.38
1.89
1.89
0.18
0.76
0.55
À4.58
À5.16
À4.95
À2.20
À3.27
À3.06
The factors limiting the PCE of polymeric solar cells include the
low exploitation of the sunlight mainly due to the narrower absorp-
tion band of the absorption spectra in the polyconjugated materials
used in the photoactive layer, in comparison with the solar spec-
trum and the mismatch of the two spectra [13]. The increased effi-
ciency of PB:PCBM solar cells compared to the reference cell made
of the normally used regioregular P3HT could be ascribed to the
stronger absorbance of PB blend in the 350–500 nm range. Our
results are very much in line with those reported in literature. In
fact, Li et al. [41] reported that a series of solar cells prepared with
polythiophene derivatives functionalized with bi(thienylenevinyl-
ene) side chains (biTV-PTs), reached a maximum PCE of 3.18%
under AM 1.5, 100 mW/cm2, with a 38% efficiency increase com-
pared to that of the cells based on P3HT under the same operative

40
M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
ciency of the photoactive layer [43]. When the organic cells are
made of a conjugated polymer and PCBM, a wider and more intense
absorption spectrum of the photoactive blend can raise the PCE of
the device acting on its photocurrent density. In our case, since
the integrated intensity of the UV–Vis spectra of P3HT, PSMe and
PB in film are in a 1:0.95:1.57 ratio respectively, and the thicknesses
of the active layers in the cells are almost the same (about 100 nm),
the higher value of JSC observed for PB (if compared with the refer-
ence cell) can be ascribed to its more extended absorption range.
To analyze the morphology of solar cells, we chose to use
atomic force microscopy (AFM). Fig. 13 shows the surface mor-
phology of P3HT or PB and PCBM (1:1 weight ratio) photoactive
blends. Films were prepared by casting the chloroform solutions
of the polymers on ITO glasses using the DB technique. The support
had been cleaned beforehand using the same procedure used for
the preparation of solar cells. After the blend deposition, samples
were subjected to the annealing procedure (30 min at 130 °C under
vacuum) and the surface images were recorded using an AFM in a
non-contact (tapping) mode in height-modulated (HMM) and
phase-modulated (PMM) modes.
Fig. 12. Band diagram with HOMO/LUMO levels of the ED polymers and EA PCBM
in relation to the work functions of ITO and Al electrodes.
conditions. The enhanced performance was attributed to the
increased absorption of (biTV-PTs) in the region between 350 and
450 nm [42]. In fact, the JSC represents the maximum photocurrent
density produced by a solar cell under solar illumination at short
circuit condition. This current is directly related to the external
quantum efficiency (EQE) which depends on the absorption effi-
The AFM images of the surfaces of the two prepared films
(100 nm thick) are quite different; in fact the surface rms (root-
mean-square) roughness is 5.4 nm for the P3HT:PCBM film (with
P3HT/PCBM
PB/PCBM
Fig. 13. AFM images of the examined blends (tapping mode, scale in nm; left: height modulated images; right: phase modulated images). The blends were obtained by
casting the polymer/PCBM solutions in CHCl3 on ITO glasses.

M. Lanzi et al. / Reactive & Functional Polymers 83 (2014) 33–41
41
an average 18 nm diameter of grains) and 2.2 nm for PB:PCBM
References
(average 11 nm diameter grains). The large black features in
PMM, which correspond to bumps in topography (HMM), are
ascribable to PCBM-rich domains. The rough surface indicates a
P3HT self-organization in the blend, enhancing the ordered struc-
ture formation in the film [44]. However, an excessively rough sur-
face morphology can cause poorer contact between the
photoactive layer and the metallic cathode [45], leading to a
decrease in the Voc values as the shunt resistance of the cell
increases [46]. Surface roughness was found to be a critical param-
eter also in polymer solar cells based on inkjet-printed PEDOT:PSS
layers, where irregular surface morphologies of the layers led to
very poor device performance in terms of Jsc, FF and PCE [47].
All devices showed good reproducibility and stability in time in
terms of final PCE: the cells stored in vacuo, at room temperature,
showed an average approximate 5% PCE decrease in the initial
value after 30 days. The best solar cell performance was obtained
using the PB/PCBM system as the active blend. This can be ascribed
not only to the enhanced absorption spectrum of this blend, which
acts positively on its generated photocurrent, but also to its very
smooth surface and to the nanoscale homogeneous separation of
its components, leading to a high interfacial area of the electron
donor–acceptor domains, which acts positively on the final Voc
and FF, as confirmed by the data reported in Table 1. The obtained
results are very promising since they demonstrate that it is possi-
ble to produce highly efficient active layers of polymer solar cells
with good sunlight absorption by using a simple mixture of easily
synthesizable polymers, thus avoiding the long and complex syn-
thesis usually required to obtain low-bandgap conjugated poly-
mers [48].
[1] K. Norrman, A. Ghanbari Siahkali, N.B. Larsen, Annu. Rep. Prog. Chem. Sect. C.
101 (2005) 174–201.
[2] P. Schilinsky, C. Waldauf, C.J. Brabec, Adv. Funct. Mater. 16 (2006) 1669–1672.
[3] F.C. Krebs, M. Jorgensen, K. Norman, O. Hagemann, J. Alstrup, T.D. Nielsen, J.
Fyenbo, K. Larsen, J. Kristensen, Sol. Energy Mater. Sol. Cells 93 (2009) 422–
441.
[4] A. Lange, M. Wegener, C. Boeffel, B. Fischer, A. Wedel, D. Neher, Sol. Energy
Mater. Sol. Cells 94 (2010) 1816–1821.
[5] F.C. Krebs, Sol. Energy Mater. Sol. Cells 93 (2009) 465–475.
[6] H.K. Roth, K. Eidner, H. Roth, Circ. World 22 (2) (1996) 31–32.
[7] R. Baumann, U. Kulish, M. Schroedner, H.K. Roth, D. Sebastian, J. Bargon, Synth.
Met. 55–56 (1993) 3643–3648.
[8] M. Lanzi, F.P. Di Nicola, M. Livi, L. Paganin, F. Cappelli, F. Pierini, J. Mater. Sci. 48
(2013) 3877–3893.
[9] H. Hoppe, M. Niggemann, C. Winder, J. Kraut, R. Hiesgen, A. Hinsch, D.
Meissner, N.S. Sariciftci, Adv. Funct. Mater. 14 (2004) 1005–1011.
[10] R. Hegde, N. Henry, B. Whittle, H. Zang, B. Hu, J. Chen, K. Xiao, M. Dadmun, Sol.
Energy Mater. Sol. Cells 107 (2012) 112–124.
[11] A.J. Moulé, S. Allard, N.M. Kronenberg, A. Tsumi, U. Scherf, K. Meerholz, J. Phys.
Chem. C 112 (2008) 12583–12589.
[12] J. Hou, H.Y. Chen, S. Zhang, G. Li, Y. Yang, J. Am. Chem. Soc. 130 (2008) 16144–
16145.
[13] Y. Li, Y. Zou, Adv. Mater. 20 (2008) 2952–2958.
[14] E. Bundgaard, F.C. Krebs, Sol. Energy Mater. Sol. Cells 91 (2007) 954–985.
[15] D. Muehlbaher, M. Scharber, M. Morana, Z.G. Zhu, D. Waller, R. Gaudiana, C.
Brabec, Adv. Mater. 18 (2006) 2884–2889.
[16] N. Blouin, A. Michaud, M. Leclerc, Adv. Mater. 19 (2007) 2295–2300.
[17] K. Takahashi, Y. Takano, T. Yamaguchi, J. Nakamura, C. Yokoe, K. Murata, Synth.
Met. 155 (2005) 51–55.
[18] L. Angiolini, T. Benelli, V. Cocchi, M. Lanzi, E. Salatelli, React. Funct. Polym. 73
(2013) 1198–1206.
[19] H.K. Roth, H. Gruber, E. Fanghaenel, A.M. Richter, W. Hoerig, Synth. Met. 37
(1990) 151–164.
[20] B.M. Novak, E. Hegen, A. Visnanathan, L. Magde, Polym. Prepr. 31 (1990) 482–
483.
[21] I. Zergioti, M. Makrygianni, P. Dimitrakis, P. Normand, S. Chatzandroulis, Appl.
Surf. Sci. 257 (2011) 5148–5151.
[22] R.S. Lowe, P.C. Ewbank, J. Lin, L. Zhai, R.D. McCullough, Macromolecules 34
(2001) 4324–4333.
[23] M. Lanzi, L. Paganin, Eur. Polym. J. 44 (2008) 3987–3996.
[24] S.P. Singh, C. Pavan Kumar, G.D. Sharma, R. Kurchania, M.S. Roy, Adv. Funct.
Mater. 22 (2012) 4087–4095.
4. Conclusions
This work demonstrated the possibility to enhance the electro-
conductivity of some films made of thioalkylic-substituted polythi-
ophene derivatives through laser light exposure. In particular, a
blend made of regioregular poly(3-methylthio)thiophene (PTSMe)
and poly(3-hexyl)thiophene (P3HT) proved to be an optimal solu-
tion for overcoming the incomplete solubility and poor filmability
of the thioalkyl-substitued polythiophenic derivative alone. The
blend used strategically combines the plastifying properties of
the alkylthiophene polymer with the ability of the photosensitive
polythiophene to increase its electrical conductivity by means of
a fast, simple laser treatment. This way, we obtained very homoge-
neous polymeric films which were easily cast from common
organic solvents and which proved to be capable of increasing their
electrical conductivity by about 7 Â 106 times after tracing. The
final electrical conductivity of the conductive traces can be easily
modulated by working on the blend composition, laser power,
and tracing speed, thus making the preparation of integrated cir-
cuits faster and simpler. We also demonstrated that the perfor-
mance of the BHJ solar cell prepared using the blend as the
photoactive layer is higher than in the reference cell made of
P3HT alone, thanks to the wider absorption spectrum and the more
homogeneous morphology of the films obtained from P3HT/PTSMe
mixtures. These results indicate that PTSMe and, in particular, its
blends with P3HT, are promising multifunctional polymeric mate-
rials for application in electronics and in plastic solar cells.
[25] F. Bertinelli, P. Costa Bizzarri, C. Della Casa, M. Lanzi, Spectrochim. Acta A 58
(2002) 583–592.
[26] A. Mikroyannidis, P. Suresh, G.D. Sharma, Org. Electron. 11 (2010) 311–321.
[27] H.K. Roth, R. Baumann, J. Bargon, M. Schroedner, Prog. Colloid Polym. Sci. 85
(1991) 157–162.
[28] J.B. Lambert, D.A. Lightner, H.F. Shurvell, R.G. Cooks, Introduction to Organic
Spectroscopy, Macmillan Ed., New York, 1987.
[29] M. Lanzi, L. Paganin, Eur. Polym. J. 45 (2009) 1118–1126.
[30] M. Lanzi, L. Paganin, React. Funct. Polym. 70 (2010) 346–360.
[31] D. Chirvase, J. Parisi, J.C. Hummelen, V. Dyakonov, Nanotechnology 15 (2004)
1317–1323.
[32] G. Li, V. Shrotriya, Y. Yao, Y. Yang, J. Appl. Phys. 98 (2005) 043704.
[33] M. Lanzi, F.P. Di Nicola, F. Errani, L. Paganin, A. Mucci, Synth. Met. 195 (2014)
61–68.
[34] J. Roncali, Chem. Rev. 92 (1992) 711–738.
[35] A.J. Nelson, S. Glenis, A.J. Frank, J. Chem. Phys. 87 (8) (1987) 5002–5006.
[36] M. Al-Ibrahim, H. Klaus Roth, U. Zhokhavets, G. Gobsch, S. Senfuss, Sol. Energy
Mater. Sol. Cells 85 (2005) 13–20.
[37] T.M. Clarke, J.R. Durrant, Chem. Rev. 110 (2010) 6736–6767.
[38] M.C. Scharber, D. Wuhlbacher, M. Koppe, P. Denk, C. Waldhauf, A.J. Heeger, C.L.
Brabec, Adv. Mater. 18 (2006) 789–794.
[39] M. Zhang, H. Wang, C.W. Tang, Appl. Phys. Lett. 97 (2010) 143503–143506.
[40] A. Gadisa, M. Svensson, M.R. Andersson, O. Ingänas, Appl. Phys. Lett. 84 (2004)
1609–1611.
[41] J. Hou, Z. Tan, Y. Yan, Y. He, C. Yang, Y. Li, J. Am. Chem. Soc. 128 (2006) 4911–
4916.
[42] B. Carsten, F. He, H.J. Son, T. Xu, L. Yu, Chem. Rev. 111 (2011) 1493–1528.
[43] X. Yang, A. Uddin, Renew. Sustain. Energy Rev. 30 (2014) 324–336.
[44] G. Li, V. Shrotriya, J.S. Huang, Y. Yao, T. Moriatry, K. Emery, Y. Yang, Nat. Mater.
4 (2005) 864–868.
[45] G. Zhao, Y. He, Y. Li, Adv. Mater. 22 (2010) 4355–4358.
[46] A. Moliton, J.M. Nunzi, Polym. Int. 55 (2006) 583–600.
[47] S.H. Eom, S. Senthilarasu, P. Uthirakumar, S.C. Yoon, J. Lim, C. Lee, H.S. Lim, J.
Lee, S.H. Lee, Org. Electron. 10 (2009) 536–542.
Acknowledgement
[48] N. Blouin, A. Michaud, D. Gendron, S. Wakim, E. Blair, R. Neagu-Plesu, M.
Belletete, G. Durocher, Y. Tao, M. Leclerc, J. Am. Chem. Soc. 130 (2008) 732–
742.
Italian PRIN 2010–2011 is gratefully acknowledged.