Synthesis of 3,4-diphenyl-substituted poly(thienylene vinylene), low-band-gap polymers via the dithiocarbamate route

Article abstract of DOI:10.1021/ma047940e

It has been demonstrated that the dithiocarbamate precursor route is a suitable pathway towards poly(thienylene vinylene) (PTV)-type low-band-gap polymers. Two particular dithiocarbamate precursor polymers and the corresponding conjugated polymers, poly(3,4-diphenyl-2,5-thienylene vinylene) and poly(3,4-bis(4-butylphenyl)-2,5-thienylene vinylene), have been studied. The introduction of butyl side chains in the latter leads to excellent solubility in common organic solvents. Both polymers have been prepared in a straightforward manner and in good yield. The thermal conversion of the precursor polymers into the conjugated structure was studied with in situ FT-IR and UV-vis spectroscopy. Also, with the latter technique, the band gap was determined and the thermochromic effect was studied and compared with the unsubstituted PTV. The HOMO and LUMO levels of the polymers were determined from UV-vis and electrochemical measurements.

Full text of DOI:10.1021/ma047940e

Macromolecules 2005, 38, 19-26
19
Synthesis of 3,4-Diphenyl-Substituted Poly(Thienylene Vinylene),
Low-Band-Gap Polymers via the Dithiocarbamate Route
A. Henckens,†,‡ K. Colladet,† S. Fourier,† T. J. Cleij,† L. Lutsen,‡ J. Gelan,† and
D. Vanderzande*,†,‡
Limburgs Universitair Centrum, Institute for Materials Research (IMO), Universitaire Campus,
Building D, B-3590 Diepenbeek, Belgium, and IMEC, Division IMOMEC, Universitaire Campus,
Wetenschapspark 1, B-3590 Diepenbeek, Belgium
Received October 6, 2004; Revised Manuscript Received October 29, 2004
ABSTRACT: It has been demonstrated that the dithiocarbamate precursor route is a suitable pathway
towards poly(thienylene vinylene) (PTV)-type low-band-gap polymers. Two particular dithiocarbamate
precursor polymers and the corresponding conjugated polymers, poly(3,4-diphenyl-2,5-thienylene vinylene)
and poly(3,4-bis(4-butylphenyl)-2,5-thienylene vinylene), have been studied. The introduction of butyl
side chains in the latter leads to excellent solubility in common organic solvents. Both polymers have
been prepared in a straightforward manner and in good yield. The thermal conversion of the precursor
polymers into the conjugated structure was studied with in situ FT-IR and UV-vis spectroscopy. Also,
with the latter technique, the band gap was determined and the thermochromic effect was studied and
compared with the unsubstituted PTV. The HOMO and LUMO levels of the polymers were determined
from UV-vis and electrochemical measurements.
Introduction
or Gilch route,13,14 a route which inherently is charac-
terized by gel formation and the instability of the
precursor polymer. Hence, a pathway toward DP-PTV
via a more suitable route is desirable. Earlier, we
reported on the synthesis of PTV via the dithiocarbam-
ate route.7 On the basis of these successful results, the
dithiocarbamate route has been chosen as the method
of choice to obtain DP-PTV. In this article, the synthesis
of DP-PTV via this dithiocarbamate route is described.
Furthermore, three different routes toward the required
bischloromethylthiophene derivatives have been com-
pared to find the most suitable synthetic procedure to
further increase the accessibility of DP-PTV for applica-
tions.
One of the major factors causing the limited process-
ability of PTV is the general insolubility of this type of
polymers. This problem can be partially circumvented
by working via a soluble precursor polymer. One of the
first examples of the use of such an indirect method to
obtain 3,4-substituted PTV derivatives is the bis-sul-
foxide precursor route.15 In this article, we have dem-
onstrated the viability of working via a soluble precursor
polymer for DP-PTV. Notwithstanding, the synthesis of
a soluble, conjugated PTV derivative could be a sub-
stantial additional advantage. We have tackled this
problem by the introduction of additional alkyl side
chains onto the DP-PTV. Therefore, this article not only
reports on the successful synthesis of DP-PTV but also
reports on the straightforward preparation of a novel
soluble derivative of DP-PTV, poly(3,4-bis(4-butylphen-
yl)-2,5-thienylene vinylene). Both polymers have been
fully characterized, and their electronic properties have
been evaluated using optical spectroscopy and electro-
chemistry. In addition, the elimination process has been
studied in more detail using temperature-dependent
FT-IR, as well as UV-vis spectroscopy.
Since their discovery, conjugated polymers attract
continuing interest as a result of their suitability in a
broad range of application fields, such as batteries,
electroluminescent devices,1 field-effect transistors,2 and
photovoltaics.3 For the latter two applications, poly-
(thienylene vinylene) (PTV) has already proven to be
an interesting conjugated polymer with a high con-
ductivity.4-7 In addition, PTV has a band gap of about
1.7 eV and, as a result, can be regarded as a low-band-
gap polymer. Furthermore, an advantage of PTV and
its derivatives is their high absorption in the visible
range of the spectrum, making them excellent candi-
dates for photovoltaic applications.8,9 However, the
limited processability of unsubstituted PTV is trouble-
some. Hence, our current research efforts focus on
substituted PTV derivatives with improved properties.
An example of such a substituted PTV derivative is
poly(3,4-diphenyl-2,5-thienylene vinylene) (DP-PTV). It
is anticipated that, in this particular derivative, the
steric hindrance of the reactive 3 and 4 positions by the
two phenyl rings will lead to a higher stability than its
unsubstituted analogue, which is advantageous for
applications. Additionally, it is expected that the intro-
duction of phenyl substituents will lead to a decrease
in the band gap, yielding an even broader absorption of
visible light, which would increase the applicability of
this class of conjugated materials in photovoltaic ap-
plications. Despite these promising features, surpris-
ingly only limited reports exist on the synthesis and use
of DP-PTV. These few reports focus on the modification
of the optical properties of DP-PTV via near-field optical
microscopy.10-12 However, the DP-PTV employed in
those studies was synthesized via the chlorine precursor
* Author to whom correspondence should be addressed.
Phone: +32 (0)11 26 83 21. Fax: +32 (0)11 26 83 01. E-mail:
dirk.vanderzande@luc.ac.be.
Experimental Section
† Limburgs Universitair Centrum, Institute for Materials Re-
search (IMO), Universitaire Campus.
General. Unless otherwise stated, all reagents and chemi-
cals were obtained from commercial sources (Acros and Ald-
‡ IMEC, Division IMOMEC, Universitaire Campus.
10.1021/ma047940e CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/10/2004

20 Henckens et al.
Macromolecules, Vol. 38, No. 1, 2005
Scheme 1. Synthetic Route 1 toward the Synthesis of 3,4-Diphenyl-2,5-bischloromethylthiophene 5a: i) EtOH; ii)
NaOH, EtOH; iii) PhC(O)C(O)Ph; iv) LiAlH4, THF; v) SOCl2, THF
rich) and used without further purification. Tetrahydrofuran
(THF) and CH3CN were dried by distillation from sodium/
benzophenone and CaH2, respectively.
as a solid. The mixture was stirred at ambient temperature
for 3 days. After addition of water, the alcohol was evaporated.
The formed solid (benzil) was filtered off, and the filtrate was
acidified with HCl, upon which 3 precipitated. After filtra-
NMR spectra were recorded with a Varian Inova 300
1
1
spectrometer at 300 MHz for H NMR and at 75 MHz for 13C
tion, a white solid was obtained (2.02 g, 64% yield); H NMR
(CD3OD): 7.12-7.05 (m, 6H), 7.02-6.93 (m, 4H); DIP MS (CI,
NMR using a 5 mm probe. Gas chromatography/mass spec-
trometry (GC/MS) analyses were carried out with TSQ-70 and
Voyager mass spectrometers (Thermoquest); the capillary
column was a Chrompack Cpsil5CB or Cpsil8CB. Analytical
size exclusion chromatography (SEC) was performed using a
Spectra series P100 (Spectra Physics) pump equipped with
two mixed-B columns (10 µm, 2 cm × 30 cm, Polymer Labs)
and a refractive index detector (Shodex) at 70°C. A DMF
solution of oxalic acid (1.1 × 10-3 M) was used as the eluent
at a flow rate of 1.0 mL/min. Molecular weight distributions
are given relative to polystyrene standards. FT-IR spectra were
collected with a Perkin-Elmer Spectrum One FT-IR spectrom-
eter (nominal resolution 4 cm-1, summation of 16 scans).
UV-vis spectroscopy was performed on a VARIAN CARY 500
UV-vis-NIR spectrophotometer (scan rate: 600 nm/min).
Samples for temperature-dependent thin-film FT-IR and
UV-vis characterization were prepared by spin-coating the
precursor polymer from a CHCl3 solution (10 mg/mL) onto
NaCl disks at 500 rpm or quartz disks at 700 rpm. The disks
were heated in a Harrick high-temperature cell (heating
rate: 2 °C/min), which was positioned in the beam of either
the FT-IR or the UV-vis spectrometer to allow in-situ mea-
surements. Spectra were taken continuously under a continu-
ous flow of N2 during which the samples were in direct contact
with the heating element. Fluorescence spectra were obtained
with a Perkin-Elmer LS-5B luminescence spectrometer.
Thin-film electrochemical properties were measured with
an Eco Chemie Autolab PGSTAT 20 potentiostat/galvanostat
using a conventional three-electrode cell (electrolyte: 0.1 mol/L
LiClO4 in anhydrous CH3CN) with an Ag/Ag+ reference
electrode (0.01 mol/L AgNO3, 0.1 mol/L LiClO4 and CH3CN),
a platinum counter electrode, and an indium-tin oxide (ITO)
coated glass substrate as working electrode. Cyclic voltammo-
grams were recorded at 50 mV/s under N2 atmosphere, and
all potentials were referenced using a known standard (fer-
rocene/ferrocinium, 0.099 V vs Ag/Ag+). Precursor polymer 13a
was spin-coated from a CHCl3 solution (5 mg/mL) onto the ITO
working electrode and subsequently converted to the conju-
gated polymer 14a by thermal treatment under a continuous
flow of N2. Polymer 14b was directly spin-coated from a CHCl3
solution (5 mg/mL) onto the ITO working electrode. Solution
electrochemistry was performed using the same setup (elec-
trolyte: 0.1 mol/L TBAPF6 in anhydrous CH2Cl2) and a Pt
working electrode.
m/e): 325 (M+ + 1), 281 (M+ + 1 - COO).
3,4-Diphenyl-2,5-bischloromethylthiophene (5a) via Method
1 (cf. Scheme 1). A solution of 3 (1 g, 3.09 mmol) in THF (20
mL) was added to a suspension of LiAlH4 (0.35 g, 9.26 mmol)
in THF (10 mL) at 0 °C. The mixture was allowed to warm to
ambient temperature, was stirred for 2 h, and then was heated
to reflux temperature for an additional 2 h. The reaction
mixture was quenched by the addition of HCl (1 M) until the
reaction mixture was neutral. After extraction with CHCl3 (3
× 20 mL), the combined organic layers were dried over MgSO4
and the solvent was evaporated. The alcohol formed in this
way was used without further purification [DIP MS (EI, m/e):
296 (M+), 279 (M+ - OH), 249]. To a solution of the alcohol in
THF (5 mL), thionyl chloride (0.5 mL, 6.790 mmol) was added
dropwise at 0°C. The mixture was stirred at room temperature
for 2 h, after which the mixture was neutralized with NaHCO3.
After extraction with CHCl3 (3 × 20 mL), the combined organic
layers were dried over MgSO4 and the solvent was evaporated
giving a white solid (430 mg, 42% yield); 1H NMR (CDCl3):
7.24-7.21 (m, 6H), 7.08-7.05 (m, 4H), 4.67 (s, 4H); MS (EI,
m/e): 332 (M+), 297 (M+ - Cl), 261 (M+ - 2 Cl). Upon
purification by column chromatography (silica, CHCl3/hexane
3:7, 6:4, 1:0, gradual enrichment of CHCl3) of 5a, some
degradation occurred as a result of its instability.
1,5-Diphenyl-3-thiapentane-1,5-dione (7). A solution of
Na2S‚9H2O (10.5 g, 43.6 mmol) in water (50 mL) was added
to a solution of 2-bromoacetophenon (15 g, 87.2 mmol) in
acetone (70 mL) over a period of 1 h at 0 °C. The mixture was
allowed to warm to ambient temperature and was stirred for
an additional hour. After dilution with water (100 mL), an
extraction with CHCl3 (3 × 50 mL) was performed and the
combined organic layers were dried over MgSO4. Evaporation
of the solvent gave a yellow solid (10.0 g, 42% yield); 1H NMR
(CDCl3): 7.97 (dd, 4H), 7.60-7.55 (tt, 2H), 7.48-7.43 (t, 4H),
3.97 (s, 4H); MS (EI, m/e): 270 (M+), 237, 165, 105.
3,4-Diphenythiolane-3,4-diol (8). To a well-stirred sus-
pension of Zn powder (7.22 g, 111 mmol) and 7 (5 g, 18.5 mmol)
in THF (60 mL) under argon atmosphere and at -18 °C, TiCl4
(6.1 mL, 55.6 mmol) was added dropwise over a period of 3 h.
The mixture was allowed to warm to 0 °C and stirred for 3 h
at 0-15 °C. The reaction was quenched by addition of crushed
ice (60 g), and the pH of the solution was brought to 9 by
adding a saturated aqueous solution of K2CO3. Dichlo-
romethane (150 mL) was added, and the mixture was stirred
overnight, after which it was filtered over a large glass filter
on which a layer of Celite was placed. The organic layer of
the filtrate was washed with water (5 × 250 mL) and dried
over MgSO4, and evaporation of the solvent gave a brown
partially crystallized solid (4.50 g, 89% yield); 1H NMR
(CDCl3): 7.19-7.05 (m, 10H), 3.62 (d, 2H), 3.15 (d, 2H); MS
(EI, m/e): 236 (M+ - H2O), 152, 120, 105.
Synthesis. Diethyl Thiodiacetate (2). A mixture of
thiodiglycolic acid (10 g, 66.7 mmol), absolute ethanol (30 mL),
and 3 drops of concentrated H2SO4 as a catalyst was allowed
to react overnight at reflux temperature. After cooling, an
extraction with CHCl3 (3 × 50 mL) was performed and the
combined organic layers were dried over MgSO4. After evapo-
ration of the solvent, a colorless liquid was obtained (12.36 g,
1
90% yield); H NMR (CDCl3): 4.17 (q, 4H), 3.60 (s, 4H), 1.26
(t, 6H); MS (EI, m/e): 206 (M+), 161 (M+ - OEt), 133 (M+
-
3,4-Diphenylthiophene (9a). Compound 9a was prepared
via two different synthetic methods (vide infra). (Method a) A
mixture of 8 (4.5 g, 16.5 mmol) and TsOH‚H2O (1.57 g, 8.27
mmol) in benzene (30 mL) was heated at reflux temperature
for 4 h during which the azeotropically formed water was
COOEt), 104 (M+ - COOEt - Et).
3,4-Diphenylthiophene-2,5-dicarboxylic Acid (3). To a
mixture of 2 (2 g, 9.71 mmol) and benzil (2.04 g, 9.71 mmol)
in methanol (15 mL), NaOMe (1.57 g, 29.1 mmol) was added

Macromolecules, Vol. 38, No. 1, 2005
Synthesis of Low-Band-Gap Polymers 21
Table 1. Polymerization Results for Monomers 12a (entry
Scheme 2. Synthetic Routes 2 and 3 toward the
Synthesis of 3,4-Diphenyl-2,5-bischloromethyl-
thiophene 5a, as Well as the Synthesis of the
Dithiocarbamate Monomer 12a and Polymers 13a and
14a: i) Na2S‚9H2O, Me2CO/H2O; ii) TiCl4, Zn, THF; iii)
TsOH‚H2O, PhH; iv) Pd(PPh3)4, KF, Toluene/H2O; v)
CH2O, HCl, Ac2O; vi) NaSC(S)NEt2‚3H2O, MeOH; vii)
LDA, THF; viii) ∆T
1-3) and 12b (entry 4)
polymerization
entry temperature (°C) yield (%) Mw (× 10-3
)
PD
1
2
3
4
0
-78
20
40
35
38
50.4
1.4
1.2
29.8
-78 f 0
-78
24.6
1.2
16.2, 3.3a
1.2, 1.1a
a Bimodal molecular weight distribution.
141.14 (2C), 135.33 (2C), 133.39 (2C), 129.99 (4C), 127.82 (4C),
126.77 (2C), 49.31 (2C), 46.63 (2C), 35.82 (2C), 12.39 (2C),
11.43 (2C).
3,4-Bis(4-butylphenyl)thiophene (9b). Compound 9b
was obtained in an analogous way to that described for 9a
(method b), starting from 4-butylphenylboronic acid (2.5 g, 14.0
mmol) and 3,4-dibromothiophene (0.885 g, 3.66 mmol). A
colorless oil was obtained (730 mg, 60% yield); 1H NMR
(CDCl3): 7.26 (s, 2H), 7.08 (m, 8H), 2.58 (t, 4H), 1.59 (m, 4H),
1.35 (m, 4H), 0.92 (t, 6H); MS (EI, m/e): 348 (M+), 262 (M+
2C3H7), 248 (M+ - C3H7 - C4H9), 234 (M+ - 2C4H9).
-
3,4-Bis(4-butylphenyl)thiophene-2,5-diylbismethylene-
N,N-diethyldithiocarbamate (12b). Compound 12b was
obtained in an analogous way to that described for 12a.
Starting from 9b (730 mg, 2.09 mmol), paraformaldehyde (170
mg, 5.66 mmol), concentrated HCl (1.18 g, 12.0 mmol), and
acetic anhydride (2.14 g, 20.9 mmol), 5b was obtained. Without
further isolation, 5b was reacted with sodium diethyldithio-
carbamate trihydrate (1.09 g, 4.84 mmol). The dithiocarbamate
monomer 12b was obtained as an orange oil (900 mg, 64%
1
yield); H NMR (CDCl3): 7.01-6.94 (AA′BB′, 4H), 6.93-6.86
(AA′BB′, 4H), 4.63 (s, 4H), 4.01 (q, J ) 7.2 Hz, 4H), 3.71 (q, J
) 7.2 Hz, 4H), 2.52 (t, J ) 7.2 Hz, 4H), 1.53 (q, J ) 7.2 Hz,
4H), 1.27 (2t, J ) 7.2 Hz, 12H), 0.88 (t, J ) 7.2 Hz, 6H); 13C
NMR (CDCl3): 194.21 (2C), 141.24 (2C), 141.11 (2C), 132.85
(2C), 132.54 (2C), 129.76 (4C), 127.73 (4C), 49.23 (2C), 46.58
(2C), 35.93 (2C), 35.08 (2C), 33.11 (2C), 22.07 (2C), 13.78 (2C),
12.36 (2C), 11.41 (2C).
removed using a Dean-Stark apparatus. The obtained mix-
ture was washed with an aqueous solution of NaHCO3 (5%,
30 mL) and water (30 mL), after which it was dried over
MgSO4 and the solvents were removed by evaporation. The
crude mixture was purified by column chromatography (silica,
CHCl3/n-hexane 1:4), and a white solid was obtained (3.0 g,
77% yield).
Polymerization of 12a (13a). The polymerization of 12a
was performed using three different reaction conditions (cf.
entries 1-3, Table 1). A solution of monomer 12a (400 mg,
0.716 mmol) in dry THF (3.6 mL, 0.2 M) at -78 °C (entries 2
and 3) or 0 °C (entry 1) was degassed for 15 min by passing
through a continuous stream of N2, after which LDA (360 µL
of a 2 M solution in THF/n-heptane) was added in one portion.
The mixture was kept at -78 °C (entries 2 and 3) or 0 °C (entry
1) for 90 min, during which time the passing of N2 was
continued. Subsequently, depending on the polymerization
procedure, ethanol (6 mL) was added at -78 °C to stop the
reaction (entry 2) or the solution was allowed to come to 0 °C
(entry 3). The reaction mixture was quenched in ice water (100
mL), after which it was neutralized with HCl (1 M in H2O).
Subsequently, the aqueous phase was extracted with CH2Cl2
(3 × 60 mL). The organic layers were combined, after which
the solvents were removed by evaporation under reduced
pressure. The resulting crude polymer was redissolved in
CHCl3 (2 mL), and a precipitation was performed in MeOH
(100 mL) at 0 °C. The polymer was collected and dried in vacuo
and further purified to remove traces of monomer by repre-
cipitation from diethyl ether/n-hexane (3:1; 100 mL) at 0 °C.
A white solid was obtained (59-117 mg, 20-40% yield). 1H
NMR (CD2Cl2): 7.21-6.94 (br m, 6H), 6.84-6.60 (br m, 4H),
5.20-5.10 (br s, 1H), 4.10-3.85 (br s, 2H), 3.85-3.55 (br s,
2H), 3.55-3.10 (br s, 2H), 1,42-1.13 (br s, 6H); 13C NMR
(CD2Cl2): 193.40, 140.48, 139.84, 137.81, 136.33 (2C), 135.76,
130.43 (4C), 127.88 (4C), 126.62 (2C), 51.63, 49.25, 47.03,
38.25, 12.89, 11.63. The residual fractions contained low-
molecular-weight oligomers and monomer residues.
(Method b) Phenylboronic acid (5 g, 41.0 mmol), 3,4-
dibromothiophene (2.14 g, 8.85 mmol), and KF (2.10 g, 36.2
mmol) were dissolved in water (40 mL) and toluene (40 mL).
Pd(PPh3)4 (730 mg) was added as a catalyst. After the mixture
was refluxed for 18 h, an extraction with CHCl3 (3 × 50 mL)
was performed and the combined organic phases were dried
over MgSO4. The crude reaction product was purified by
column chromatography (silica, n-hexane), and a white solid
was obtained (1.6 g, 77% yield); 1H NMR (CDCl3): 7.30 (s, 2H),
7.25-7.21 (m, 6H), 7.19-7.16 (m, 4H); MS (EI, m/e): 236 (M+).
3,4-Diphenyl-2,5-bischloromethylthiophene (5a) via Methods
2 and 3 (cf. Scheme 2). To a mixture of paraformaldehyde (343
mg, 11.4 mmol) and 9a (1 g, 4.24 mmol), concentrated HCl
(2.35 g, 24.2 mmol) and acetic anhydride (4.32 g, 42.4 mmol)
were added dropwise under N2 atmosphere at 0 °C. The
mixture was heated at 70 °C for 4.5 h, after which a cold (0
°C) saturated aqueous solution of sodium acetate (10 mL) and
a 25% aqueous solution of sodium hydroxide (10 mL) were
added. The mixture was extracted with CHCl3 (3 × 50 mL)
and dried over MgSO4. After evaporation of the solvents, a
1
light brown solid was obtained (1.38 g, 98% yield); H NMR
(CDCl3): 7.24-7.21 (m, 6H), 7.08-7.05 (m, 4H), 4.67 (s, 4H);
MS (EI, m/e): 332 (M+), 297 (M+ - Cl), 261 (M+ - 2 Cl).
3,4-Diphenylthiophene-2,5-diylbismethylene-N,N-di-
ethyldithiocarbamate (12a). A mixture of 5a (3 g, 8.89
mmol) and sodium diethyldithiocarbamate trihydrate (4.6 g,
20.4 mmol) in 10 mL of methanol was stirred for 3 h at
ambient temperature. Subsequently, the mixture was ex-
tracted with diethyl ether (3 × 50 mL) and dried over MgSO4.
The crude reaction mixture was purified by column chroma-
tography (silica, n-hexane/CHCl3 3:7), after which the dithio-
carbamate monomer was obtained as a white solid (3.47 g, 70%
yield); 1H NMR (CDCl3): 7.24-7.14 (m, 6H), 7.03-7.00 (m,
4H), 4.62 (s, 4H), 4.01 (q, J ) 7.2 Hz, 4H), 3.69 (q, J ) 7.2 Hz,
4H), 1.26 (2t, J ) 7.2 Hz, 12H); 13C NMR (CDCl3): 194.22 (2C),
Polymerization of 12b (13b). Compound 13b was ob-
tained in an analogous way to that described for 13a starting
from 12b (575 mg, 0.857 mmol) and LDA (429 µL of a 2 M
solution in THF/n-heptane). A white solid was obtained (171
mg, 38% yield). 1H NMR (CD2Cl2): 7.01-6.76 (br m, 4H), 6.76-
6.53 (br m, 4H), 5.51-5.15 (br s, 1H), 4.14-3.84 (br s, 2H),

22 Henckens et al.
Macromolecules, Vol. 38, No. 1, 2005
Scheme 3. Synthesis of the Butyl-Substituted 3,4-Diphenylthiophene Monomer 12b and Polymers 13b and 14b:
iv) Pd(PPh3)4, KF, Toluene/H2O; v) CH2O, HCl, Ac2O; vi) NaSC(S)NEt2‚3H2O, MeOH; vii) LDA, THF; viii) ∆T
kayama et al.18,19 The latter involves the synthesis of
3.84-3.54 (br s, 2H), 3.42-3.16 (br s, 2H), 2.63-2.39 (br t,
4H), 1.64-1.40 (br m, 4H), 1,40-1.07 (br m, 10H), 0.99-0.78
(br t, 6H); 13C NMR (CD2Cl2): 193.82, 141.48, 140.90 (3C),
137.31, 136.02, 133.78 (2C), 130.45 (4C), 127.81 (4C), 51.87,
49.13, 47.06, 38.30, 35.62 (2C), 33.75 (2C), 22.66 (2C), 14.08
(2C), 12.89, 11.63.
thiophenes with two substituents on the 3 and 4 posi-
tions. Starting from commercially available 2-bromo-
acetophenon 6, the reaction with sodium sulfide non-
ahydrate (Na2S‚9H2O) yields 1,5-diphenyl-3-thiapen-
tane-15-dione 7. Subsequently, 3,4-diphenylthiolane-3,4-
diol 8 forms by the intramolecular reductive coupling
reaction with a low-valent titanium reagent prepared
from titanium(IV) chloride (TiCl4) and Zn powder.
Treatment of the diol with p-toluenesulfonic acid
(TsOH‚H2O) in benzene at reflux temperature gives 3,4-
diphenylthiophene 9a.
Thermal Conversion of 13b in Solution (14b). A solu-
tion of 13b (200 mg) in chlorobenzene (10 mL) was heated to
150 °C and stirred for 20 h. After being cooled to 50 °C, the
resulting blue solution was precipitated dropwise in methanol
(300 mL). The polymer was filtered off and dried at room
temperature under reduced pressure. A black solid was
obtained (140 mg, 95% yield); 1H NMR (CDCl3): 7.07-6.92
(m, 4H), 6.92-6.73 (m, 6H), 2.70-2.36 (t, 4H), 1.68-1.41 (m,
4H), 1.41-1.11 (m, 4H), 0.99-0.78 (t, 6H); 13C NMR (CDCl3):
141.88 (2C), 141.40 (2C), 136.52 (2C), 132.58 (2C), 130.41 (4C),
127.82 (4C), 121.80 (2C), 35.35 (2C), 33.30 (2C), 22.28 (2C),
13.99 (2C).
Thermal Conversion of 13a and 13b in a Thin Film
(14a and 14b). The precursor polymer 13a or 13b was spin-
coated from a CHCl3 solution (10 mg/mL) onto NaCl disks at
500 rpm or quartz disks at 700 rpm, respectively. Subse-
quently, the disks were placed in a thermo cell. A dynamic
heating rate of 2 °C/min up to 350 °C under a continuous flow
of N2 was used for the conversion process.
The third route (Scheme 2, steps iv and v) also
involves the synthesis of 9a. However, in this route, 9a
is directly synthesized via a Suzuki coupling reaction
between the commercially available phenylboronic acid
10a and 3,4-dibromothiophene 11. A comparison of the
two methods to obtain 9a reveals that the Suzuki
coupling reaction is preferred since it requires only one
step and the yield (77%) is better than the overall yield
(29%) of the three step synthesis. Having obtained 9a
via the Suzuki coupling, the desired dichloride 5a is
prepared via a typical chloromethylation reaction.20 It
should be noted that 5a prepared via this route does
not require purification by column chromatography,
which can lead to degradation (vide supra). Instead, the
dithiocarbamate monomer 12a forms directly from the
bischloromethyl compound 5a by reaction with sodium
diethyldithiocarbamate trihydrate.
In view of the better overall monomer yields, the
straightforward preparation, and conversion of 5a with-
out degradation, as well as the lowest number of
reaction steps, the third route involving the Suzuki
coupling of the phenyl with the thiophene ring has been
chosen to prepare 12a. The overall yield from the
starting compounds toward 12a is 53%. Similar consid-
erations have motivated our choice to prepare the butyl-
substituted monomer 12b only via this route (Scheme
3). This synthesis is straightforward, and the overall
yield from the starting compounds toward 12b is 39%.
Polymerization. The polymerization of monomer
12a has been performed using three different reaction
conditions (cf. entries 1-3, Table 1), deviating in the
temperatures employed. For monomer 12b, only one
procedure has been selected (cf. entry 4, Table 1). For
all reactions, lithium diisopropyl amide (LDA) is used
as a base and dried THF as the solvent (Schemes 2 and
3). After the polymerization was allowed to proceed for
90 min, termination of the reaction is achieved by
Results and Discussion
Monomer Synthesis. In view of the advantages
compared to the Gilch route, the dithiocarbamate
precursor route has been chosen as the precursor
method of choice to obtain the PTV derivatives (vide
supra). The monomers used in the dithiocarbamate
precursor route are synthesized from a dihalogenide in
a single-step reaction.7 Three different synthetic routes
toward the required dihalogenide, i.e. 3,4-diphenyl-2,5-
bischloromethylthiophene 5a, have been successfully
tested. The first route (Scheme 1) starts with the
esterification of the carboxylic groups of commercially
available thiodiglycolic acid 1 with ethanol.16 Next, the
Hinsberg reaction between diethyl thiodiacetate 2 and
benzil gives the thiophene ring system.17 Consecutive
reduction of the two carboxylic groups of 3 with lithium
aluminum hydride (LiAlH4) followed by chlorination of
the diol with thionyl chloride (SOCl2) yields the desired
dichloride 5a. Upon purification by column chromatog-
raphy (cf. Experimental Section) of 5a prepared via this
first route, some degradation occurred as a result of its
instability.
In the second route (Scheme 2, steps i, ii, iii, and v)
toward dichloride 5a, 3,4-diphenylthiophene 9a has first
been synthesized via the method developed by Na-

Macromolecules, Vol. 38, No. 1, 2005
Synthesis of Low-Band-Gap Polymers 23
pouring the polymerization mixture in ice water fol-
lowed by neutralization with hydrochloric acid to pH )
7. To avoid possible side reactions, the polymerizations
performed at -78 °C are first quenched with ethanol
at -78 °C prior to this acidification step (only entries 2
and 4, Table 1). After extraction, the polymers are
purified using reprecipitation. The isolated yields range
from 20 to 40% (Table 1), with the highest yields being
associated with the low polymerization temperatures.
The weight-average molecular weights (Mw) of 13a
and 13b have been determined by analytical SEC in
DMF using polystyrene standards. The observed mo-
lecular-weight distributions for 13a are monomodal,
with Mw values ranging from 24.6 × 103 to 50.4 × 103
and low polydispersities (cf. entries 1-3, Table 1). In
contrast, polymer 13b exhibits a bimodal weight dis-
tribution with a lower apparent Mw (Mw ) 16.2 × 103
and 3.3 × 103; cf. entry 4, Table 1). However, it should
be noted that since all molecular weights are referenced
to polystyrene standards, the actual value for Mw will
differ from the apparent Mw observed with analytical
SEC.
Figure 1. Temperature-dependent UV-vis spectra of the
elimination of 13a giving 14a.
The electrochemical properties of 13a and 13b are
typical for nonconjugated polymers, with both the oxi-
dation and the reduction being irreversible. The onset
of oxidation in solution is observed at ca. 0.95 V vs
Ag/Ag+ (-5.7 eV) and is associated with the oxidation
of the thiophene groups. The onset of reduction is
observed at ca. -1.3 V vs Ag/Ag+ (-3.5 eV) and is
associated with the dithiocarbamate groups, as well as
with impurities.
Thermal Conversion. The synthesized precursor
polymers can be transferred into the conjugated polymer
by a thermal treatment. Upon heating, the dithiocar-
bamate group of polymer 13a,b is eliminated to form
the corresponding conjugated polymer 14a,b (Schemes
2 and 3). The thermal conversion can be performed in
a thin film or in solution, depending on the type and
solubility of the polymer used. Thin-film processes have
the added advantage that the thermal elimination
reaction and the thermal stability of the polymer can
be followed by means of in-situ UV-vis and FT-IR
spectroscopy. In this way, one can deduce a reproducible
protocol to convert the precursor into the conjugated
form, which is essential for applications. The thin-film
conversion process was used to obtain both 14a and 14b.
In contrast to the thin-film elimination, the solution
elimination process can be accomplished on a large
scale, although this is only of interest when the conju-
gated polymer obtained is soluble. Hence, the solution
elimination was only performed to obtain the butyl-
substituted DP-PTV 14b.
Figure 2. UV-vis absorbance profiles at 247, 403, and 590
nm as a function of temperature during the elimination of 13a.
Formation of 14a in a Thin Film. Upon heating a
thin film of 13a, the gradual formation of the con-
jugated structure of polymer 14a is observed. This is
well visible in the temperature-dependent UV-vis
absorption spectrum by the development of a band with
λmax ) 550-600 nm and the decrease of the absorption
of 13a at λmax ) 247 nm (Figure 1). The thermal process
can be better analyzed using the absorbance profiles
(Figure 2). In these profiles, an increase in the absor-
bance at 590 nm can be seen between 110 and 160 °C,
reflecting the formation of the conjugated system.
Furthermore, in the region between 80 and 160 °C, the
formation and disappearance of partially conjugated
segments (λmax ) 403 nm) is visible. The conversion
process is also readily observed using temperature-
Figure 3. Temperature-dependent FT-IR spectra of the
elimination of 13a giving 14a.
dependent FT-IR spectroscopy (Figure 3). The absorp-
tion bands at 1486, 1416, 1267, and 1206 cm-1 all arise
from the dithiocarbamate group. From the profiles in
Figure 4, it is clear that the increase in double bond
absorption (934 cm-1) starts just above 100 °C and the

24 Henckens et al.
Macromolecules, Vol. 38, No. 1, 2005
Figure 4. IR absorption profiles at 934 and 1267 cm-1 as a
function of temperature during the elimination of 13a.
Figure 5. Temperature-dependent UV-vis spectra of the
elimination of 13b giving 14b.
conversion process is complete around 130 °C for the
applied temperature program (heating rate at 2 °C/min).
This small difference, as compared to the temperature-
dependent UV-vis data, is a result of the increased
sensitivity of FT-IR, viz. any double bond formed will
be visible in FT-IR, whereas only longer conjugated
systems will be visible in the UV-vis. Hence, temper-
ature-dependent FT-IR spectroscopy is an excellent
method to study the start and nature of the elimination
process, whereas temperature-dependent UV-vis spec-
troscopy is the most suitable method to determine the
end-point of the elimination process. It is noteworthy
that the thermal stability of 14a is 300 °C. This is higher
than unsubstituted PTV, which already displays around
275 °C a notable decrease in intensity of the UV-vis
absorption maximum.7
The room-temperature absorption maximum λmax of
a thin film of 14a is 600 nm. This is notably higher than
λmax observed by previous researchers for thin films of
the same polymer (i.e., 54613 and 540 nm14). These
anomalous results are possibly the result of either the
presence of polymer defects or the occurrence of incom-
plete conversion in the previous studies; viz. only
comparatively short conjugated systems were present.
This is in agreement with our observations in the
temperature-dependent UV-vis measurements (Figure
1), in which the λmax of a thin film of 14a gradually shifts
to higher wavelength upon increase of temperature. In
our experiments, all films are heated until no increase
of λmax occurs, i.e., full conjugation is achieved. Upon
heating 14a, a blue-shift of λmax is observed as a result
of the thermochromic effect (175 °C: λmax ) 590 nm).
This thermochromic shift of 10 nm is smaller than the
shift observed for unsubstituted PTV (i.e., 25 nm),
possibly reflecting the reduced mobility of the phenyl-
substituted PTV main chain.7 The band gap of 14a at
room temperature can be derived from its UV-vis
characteristics by drawing the tangent on the low
energetic edge of the absorption spectrum. The intersec-
tion with the abscissa is at 730 nm, which corresponds
to a band gap of 1.70 eV, which is almost identical to
the band gap observed for PTV, i.e., 1.67 eV.7 Appar-
ently, no reduction in band gap is observed as a result
of phenyl substitution, possibly as a result of competing
steric effects. In agreement to unsubstituted PTV, as
well as other substituted PTV derivatives,21 thin films
Figure 6. UV-vis absorbance profiles at 248, 391, and 550
nm as a function of temperature during the elimination of 13b.
and solutions of polymer 14a do not exhibit photolumi-
nescence at room temperature. This is in contrast to
copolymers containing PTV repeating units, which do
exhibit photoluminescence, although with a low quan-
tum yield.22
Formation of 14b in a Thin Film. Precursor
polymer 13b was converted in a thin film into 14b under
similar conditions as 13a. Temperature-dependent
UV-vis and FT-IR measurements (Figures 5-8) indi-
cate that the introduction of butyl groups surprisingly
has only a small impact on the elimination process, i.e.
the elimination process starts at ca. 90 °C. However, it
should be noted that the profiles (Figures 6 and 8)
indicate that degradation of the conjugated system
commences around 275 °C. Apparently, the introduction
of the butyl side chains somewhat reduces the thermal
stability. Notwithstanding, thermal stability up to 275
°C is sufficiently high for virtually all applications.
The introduction of alkyl side chains has also an
impact on the UV-vis absorption properties. The room-
temperature absorption maximum, λmax, of a thin film
of 14b is 590 nm, which is slightly blue-shifted with
respect to the absorption maximum observed for 14a.
This shift is also observed for the band gap. The low

Macromolecules, Vol. 38, No. 1, 2005
Synthesis of Low-Band-Gap Polymers 25
Figure 9. UV-vis absorption spectra of 14b in a CHCl3
solution (solid) and as a thin film spin-coated from a CHCl3
solution (dashed).
Figure 7. Temperature-dependent FT-IR spectra of the
elimination of 13b giving 14b.
The apparent molecular weight (Mw) observed with
analytical SEC of 14b prepared in this way is 9.0 × 103
(solvent, THF; polydispersity, PD ) 1.7). The observed
absorption maximum (λmax) of a thin film of 14b spin-
coated from a CHCl3 solution is 592 nm. The UV-vis
absorption spectrum of a solution of 14b in CHCl3
exhibits an absorption maximum λmax ) 577 nm (Figure
9). This is 15 nm blue-shifted with respect to the absorp-
tion maximum of a thin film of 14b as a result of
electronic interchain interactions in the solid state.
Similarly to 14a, solutions and thin-films polymer 14b
do not exhibit photoluminescence at room temperature
(vide supra). The stability of spin-coated thin films of
14b formed in solution was investigated using temper-
ature-dependent UV-vis spectroscopy. The onset of
degradation occurs around 180 °C, as compared to 275
°C in the films of 14b formed directly in a thin film.
The origin of this difference is not certain but may be
related to the exposure of the solution-formed 14b to
ambient conditions after synthesis and during spin-
coating.
Figure 8. IR absorption profiles at 936 and 1266 cm-1 as a
function of temperature during the elimination of 13b.
Electrochemistry. Cyclic voltammetry (CV) was
employed to investigate the electrochemical behavior of
polymers 14a and 14b and to estimate their highest
occupied molecular orbital (HOMO) energy levels. The
cyclic voltammograms of 14a and 14b display distinct
oxidation and reduction processes (Figure 10). Whereas
the oxidation process is directly associated with the
conjugated structure, i.e. p-doping, the reduction process
is irreversible, poorly defined, and associated with
polymer defects and traces of impurities. A similar
reduction behavior was observed for the precursor
polymers 13a and 13b. Hence, no n-doping has been
observed. In contrast, the observed p-doping process is
fully reversible. The oxidation potential of 14a is 0.41
V vs Ag/Ag+. However, the HOMO energy level is more
accurately determined from the onset of oxidation. In
the anodic scan, the onset of oxidation of 14a occurs at
0.19 V vs Ag/Ag+, which corresponds to a HOMO energy
level of -4.89 eV. The p-doping for 14b is also reversible
(oxidation potential 0.46 V vs Ag/Ag+), although not as
well defined as for 14a. The onset of oxidation of 14b
occurs at 0.30 V vs Ag/Ag+, which corresponds to a
HOMO energy level of -5.00 eV. For comparison, the
oxidation potential of PTV is 0.39 V vs Ag/Ag+ with an
onset of oxidation at 0.08 V vs Ag/Ag+, which corre-
sponds to a HOMO energy level of -4.78 eV. The shift
energetic edge of the absorption spectrum of 14b is at
701 nm, which corresponds to a band gap of 1.77 eV.
This increase in band gap, as compared to 14a, is oppo-
site to what can be expected for the introduction of
lightly electron donating substituents. Apparently, the
effect of the reduced electronic interchain interactions,
as well as the increased disorder in the conjugated
system created by the introduction of the alkyl side
chains, is larger than the possible electronic substituent
effects.
Formation of 14b in Solution. In addition to the
thin-film conversion (vide supra), highly soluble 13b has
also been converted into 14b in solution. Chlorobenzene
was used as the solvent, and the process was heated to
150 °C (cf. Experimental Section). The solution conver-
sion process is virtually quantitative (yield: 95%). The
optical properties of 14b prepared in solution are
identical to those found for 14b prepared in the thin-
film process. It is noteworthy that 14b is soluble in a
wide variety of common organic solvents, including
CHCl3, CH2Cl2, THF, toluene, and chlorobenzene. Hence,
14b can be analyzed in solution, as well as spin-coated
from a solution, thus considerably increasing the ap-
plicability of this polymer for electronic applications.

26 Henckens et al.
Macromolecules, Vol. 38, No. 1, 2005
strated. The monomer preparation procedures have
been optimized and DP-PTV 14a is readily accessible.
As a result of the introduction of phenyl substituents,
14a demonstrates an enhanced thermal stability as
compared to PTV. Furthermore, 14a exhibits a signifi-
cantly improved λmax value as compared to previous
reports, indicating its improved purity and lower defect
levels. In addition to DP-PTV, a novel dibutyl-substi-
tuted DP-PTV 14b has been prepared, which is soluble
in common organic solvents. This allows the solution
processing of DP-PTV derivatives for applications. Both
polymers have a low band gap (1.7-1.8 eV) and can be
reversibly p-doped.
Acknowledgment. The authors gratefully acknowl-
edge The Fund for Scientific Research-Flanders (FWO)
and the Belgian Program on Interuniversity Attraction
Poles (IUAP), initiated by the Belgian State Prime
Minister’s Office, for the financial support. We also want
to thank the IWT (Institute for the Promotion of
Innovation by Science and Technology in Flanders) for
the financial support via the SBO-project 030220 “Nano-
solar”. K.C. wishes to thank the IUAP-V and S.F. the
BOF-LUC for granting a Ph.D. fellowship.
Figure 10. Cyclic voltammograms of the oxidation and reduc-
tion behavior of thin films of 14a (solid) and 14b (dashed).
References and Notes
(1) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem., Int.
Ed. 1998, 37, 403.
(2) Dimitrakopoulos, C. D.; Mascaro, D. J. IBM J. Res. Dev. 2001,
45, 1, 11.
(3) Nunzi, J.-M. C. R. Physique 2002, 3, 523.
(4) Fuchigami, H.; Tsumura, A.; Koezuka, H. Appl. Phys. Lett.
1993, 63, 10, 1372.
(5) Brown, A. R.; Jarrett, C. P.; de Leeuw, D.; Matters, M. Synth.
Met. 1997, 88, 37.
(6) Vandamme, L. K. J.; Feyaerts, R.; Trefan, G.; Detcheverry,
C. J. Appl. Phys. 2002, 91, 2, 719.
Figure 11. Cyclic voltammogram of the oxidation behavior
of a thin film of 14a.
(7) Henckens, A.; Lutsen, L.; Vanderzande, D.; Knipper, M.;
Manca, J.; Aernouts, T.; Poortmans, J. SPIE Proc. 2004, 26-
30 April, Strasbourg, France.
(8) Dhanabalan, A.; van Duren, J. K. J.; van Hal, P. A.; van
Dongen, J. L. J.; Janssen, R. A. J. Adv. Funct. Mater. 2001,
11, 4, 255.
(9) Winder, C.; Mu¨hlbacher, D.; Neugebauer, H.; Sariciftci, N.
S.; Brabec, C. J.; Janssen, R. A. J.; Hummelen, J. C. Mol.
Cryst. Liq. Cryst. 2002, 385, 93.
of the HOMO energy level in going from unsubstituted
14a to butyl-substituted 14b partially corresponds to
the change in band gap observed with UV-vis spec-
troscopy (vide supra). Combining of the electrochemical
data and the UV-vis characteristics gives an estimate
of the LUMO energy levels of 14a and 14b, i.e. -3.19
eV and -3.23 eV, respectively.
(10) Wei, P. K.; Hsu, J. H.; Hsieh, B. R.; Fann, W. S. Adv. Mater.
1996, 8, 7, 573.
It should be noted that the oxidation process can also
be observed visually. For both 14a and 14b, a color
change of the films from blue to light yellow was
observed upon oxidation. Subsequently, in the reverse
process upon dedoping, the color of the film returned to
blue. This observation further confirms the reversibility
of the p-doping process. It is noteworthy that, upon
application of increased bias (i.e., scanning to 1.0 V vs
Ag/Ag+), a second and a third oxidation process can be
observed for 14a (Figure 11). Similarly to the first
oxidation process, which is associated with the HOMO
energy level, the second oxidation process with an
oxidation potential of ca. 0.7 V vs Ag/Ag+ (-5.4 eV) is
reversible. This process is possibly associated with the
oxidation of the phenyl rings. In contrast, the third
oxidation process with an oxidation potential of ca. 0.9
V vs Ag/Ag+ (-5.6 eV) is irreversible and results in the
irreversible degradation of the conjugated system.
(11) Wei, P. K.; Hsu, J. H.; Fann, W. S.; Hsieh, B. R. Polym. Mater.
Sci. Engin. 1996, 75, 325.
(12) Wei, P. K.; Hsu, J. H.; Fann, W. S.; Hsieh, B. R. Synth. Met.
1997, 85, 1421.
(13) Hsieh, B. R.; Johnson, G. E.; Antoniadis, H.; McGrane, K.
M.; Stolka, M. U.S. Patent 5,558,904, 1996.
(14) Tan, C. H.; Inigo, A. R.; Hsu, J. H.; Fann, W. S.; Wei, P. K.
J. Phys. Chem. Solids 2001, 62, 1643-1654.
(15) Cheng, H.; Elsenbaumer, R. L. J. Chem. Soc., Chem. Com-
mun. 1995, 1451.
(16) Sankaran, B.; Reynolds, J. R. Macromolecules 1997, 30, 9,
2582.
(17) Hinsberg, O. Chem. Ber. 1910, 43, 901.
(18) Nakayama, J.; Hasemi, R. J. Am. Chem. Soc. 1990, 112, 5654.
(19) Nakayama, J.; Hasemi, R.; Yoshimura, K.; Sugihara, Y.;
Yamaoka, S. J. Org. Chem. 1998, 63, 4912.
(20) Becker, H.; Spreitzer, H.; Ibrom, K.; Kreuder, W. Macromol-
ecules 1999, 32, 4925.
(21) Goldini, F.; Janssen, R. A. J.; Meijer, E. W. J. Polym. Sci.,
Part A: Polym. Chem. 1999, 37, 4629.
Conclusions
(22) Chen, B.; Wu, Y.; Wang, M.; Wang, S.; Sheng, S.; Zhu, W.;
Sun, R.; Tian, H. Eur. Polym. J. 2004, 40, 6, 1183-1191.
In conclusion, the synthesis of DP-PTV derivatives via
the dithiocarbamate precursor route has been demon-
MA047940E