Transient Photoconductivity of Acceptor-Substituted Poly(3-butylthiophene)

Article abstract of DOI:10.1021/ja953762g

We have studied acceptor-substituted poly(3-butylthiophene) in an attempt to examine the role of acceptor moplecules as intrinsic charge traps under light excitations by measuring the transient photoconductivity response following pulse excitation.The specially synthesized acceptor molecule is a chemically prepared high electron affinity (HEA) monomer, (1-(4-nitrophenyl)-2-(3-thienyl)ethene.In the copolymers prepared with this acceptor monomer we have observed simultaneous decrease of the sub-nanosecond photoconductivity and increase of the slow component (t > 10 ns) photoconductivity, with increasing HEA group concentration in the substituted poly(butylthiophene)s.The slow component is attributed to a bimolecular recombination process.

Full text of DOI:10.1021/ja953762g

2980
J. Am. Chem. Soc. 1996, 118, 2980-2984
Transient Photoconductivity of Acceptor-Substituted
Poly(3-butylthiophene)
Y. Greenwald,†,‡ G. Cohen,§ J. Poplawski,† E. Ehrenfreund,†,| S. Speiser,‡ and
D. Davidov§
Contribution from the Solid State Institute, Department of Chemistry, and Department of Physics,
TechnionsIsrael Institute of Technology, Haifa 32000, Israel, and Racah Institute of Physics,
The Hebrew UniVersity, Jerusalem, Israel
ReceiVed NoVember 8, 1995X
Abstract: We have studied acceptor-substituted poly(3-butylthiophene) in an attempt to examine the role of acceptor
molecules as intrinsic charge traps under light excitations by measuring the transient photoconductivity response
following pulse excitation. The specially synthesized acceptor molecule is a chemically prepared high electron
affinity (HEA) monomer, 1-(4-nitrophenyl)-2-(3-thienyl)ethene. In the copolymers prepared with this acceptor
monomer we have observed simultaneous decrease of the sub-nanosecond photoconductivity and increase of the
slow component (t > 10 ns) photoconductivity, with increasing HEA group concentration in the substituted poly-
(butylthiophene)s. The slow component is attributed to a bimolecular recombination process.
1. Introduction
Photoconductivity (PC) in conjugated polymers has attracted
considerable attention in recent years. The transient PC signal
in conjugated polymers decays on a sub-nanosecond time scale
due to radiative as well as nonradiative recombination processes
of the charge carriers.1,2 Modified matrices of conjugated
polymers which contain charge traps in the excited state show
quenching of the charge carriers recombination.3 These modi-
fied matrices are of potential use in applications where longer
lifetimes and higher yields are required.
Figure 1. Photoactivation process of the bipolar state. The negative
charge is trapped on the HEA group, while the positive charge is
delocalized on the backbone.
New chemical and physical properties of conducting polymers
were achieved by adding an appropriate functional group to the
polymer chain.4 In some previous works the functional group
was placed as a pendant group apart from the conjugated
backbone in order to maintain a high level of conductivity, so
that spatial hindrance was minimized. Thiophene monomers,
which are polymerized at the 2,5 positions, can be substituted
at the 3,4 positions to produce a substituted conducting polymer.
Consequently, these structures are good candidates for the
purpose of modifying the electronic properties of the polymer.
In some studies, this concept was implemented by adding high
electron affinity (HEA) groups in order to change the energy
gap between the HOMO and the LUMO bands.5 More recently,
such substituted conjugated polymers were used as electron
collectors in a bilayer light emitting diode (LED) device with
increased efficiency.6 HEA groups, such as nitrobenzene, may
be used as substitutes having strong excited state electron affinity
in conjugated polymers, such as poly(3-alkylthiophene). In such
systems, a dipole moment is induced by photoexcitation, leading
to a bipolar excited state through a quasi-charge-transfer
mechanism.7 The amount of charge transfer is determined
directly by the separation of the two polar sites. Hence,
electrons will be, to some extent, localized in the vicinity of
the acceptor group, inhibiting their recombination with the
positive charge which is delocalized on the chain. As a result,
a slow decay of the photoconductivity is to be expected.
According to the multistate model8 for strongly coupled ion
pairs, there is an initial excited bipolar state in the acceptor-
substituted polymer where the charges are not fully separated
(“contact ion pair”). These bipolar pairs do not contribute to
the initial photocurrent. The bipolar neutral state is relatively
long lived and therefore allows the positive charge to hop to
another site (or chain) in the polymer, leading to fully separated
charge carriers (“solvation separated ion pair”). This process
enhances the long lived charge concentration which contributes
to the slow photoconductivity. The long lived charges recom-
bine eventually via a bimolecular recombination process. Since
the HEA group is attached to a π conjugation, perpendicular to
the polythiophene backbone (on the â site of the thiophene ring),
under illumination it is expected to behave as a trap (“acceptor”)
for electrons. The positive charges are expected to show
chainlike behavior (positive polarons or bipolarons9), as
demonstrated schematically in Figure 1.
† Solid State Institute, Technion.
‡ Department of Chemistry, Technion.
§ The Hebrew University.
| Department of Physics, Technion.
X Abstract published in AdVance ACS Abstracts, March 1, 1996.
(1) Moses, D.; Sinclair, M.; Heeger, A. J. Phys. ReV. Lett. 1987, 58,
2710.
Here we present a study of the dark conductivity and the
transient photoconductivity in the homopolymer poly(3-butylth-
(2) Bleier, H.; Donovan, K.; Friend, R. H.; Roth, S.; Rothberg, L.; Tubino,
R.; Vardeny, Z. V.; Wilson, G. Synth. Met. 1989, 28, D189.
(3) Lee, C. H.; Yu, G.; Moses, D.; Pakbaz, K.; Zhang, C.; Sariciftci, N.
S.; Heeger, A. J.; Wudl, F. Phys. ReV. B 1993, 48, 15425.
(4) Garnier, F. Angew. Chem. 1989, 101, 529.
(5) Otto, P.; Ladik, J. Synth. Met. 1990, 36, 327.
(6) Greenham, N. C.; Moratti, S. C.; Bradley, D. D. C.; Friend, R. H.;
Holmes, A. B. Nature 1993, 365, 628.
(7) Creutz, C.; Chou, M. H. Inorg. Chem. 1985, 89, 1601.
(8) Mataga, N. In Electron Transfer in Inorganic, Organic and Biological
Systems; Bolton, J. R., Mataga, N., McLendon, G., Eds.; Advances in
Chemistry Series, Vol. 228; A.C.S.: Washington, DC, 1989; p 131.
(9) Heeger, A. J.; Kivelson, S.; Schrieffer, J. R.; Su, W. P. ReV. Mod.
Phys. 1988, 60, 781.
0002-7863/96/1518-2980$12.00/0 © 1996 American Chemical Society

Transient PhotoconductiVity of Poly(3-butylthiophene)
J. Am. Chem. Soc., Vol. 118, No. 12, 1996 2981
cooled to ambient temperature, and filtrated, and the salt was washed
with DMF, followed by ether to give 9.9 g of white powder.
1-(4-Nitrophenyl)-2-(3-thienyl)ethene (4): 4-Nitrobenzaldehyde (3.4
g, 22.5 mmol) and 3 (9.9 g, 22.5 mmol) were dissolved in 63 mL of
absolute ethanol. A 208 mL solution of 0.4 M sodium ethoxide in
ethanol was added, and the precipitation of a yellow product was seen
immediately. The mixture was heated up in reflux for 2 h and then
cooled to ambient temperature. One-third of the ethanol was evapo-
rated, and water was inserted instead. The yellow product was filtrated.
In order to obtain a pure product, it was separated off by a flash
chromatography procedure (run by dichloromethane) yielding 3.015 g
of yellow powder. The yield was 58%, mp 166 °C. 1H-NMR
(CDCl3): δ ) 6.5-6.9 ppm (m, vinyl in a mixture of cis and trans,
2H), 7.1-7.4 ppm (m, thiophene, 3H), 7.6 ppm (d, J ) 9 Hz, phenyl
meta to nitro, 2H), 8.2 ppm (d, J ) 9 Hz, phenyl ortho to nitro, 2H).
IR (CHCl3): 1640, 1110, 950, 870, 1520, 1350 cm-1. MS: m/z 231,
184, 152, 141, 115, 45. High-resolution MS: m/z 231.0360 (calcd for
C12H9NO2S 231.0354). UV-vis (CH2Cl2): 352 nm.
Figure 2. General polymerization of the copolymer 3-HEA-thiophene/
3-butylthiophene. The 3-HEA-thiophene monomers are randomly
distributed.
Homo- and Copolymers: Polymerizations of the homopolymer and
the copolymers were done on 1-4 mmol scale of the monomers
dissolved in 0.1 M chloroform and 8 equiv of FeCl3. Polymerization
was carried out at 30 °C under nitrogen atmosphere until no HCl was
detected (typically for 2-3 h). The black solid was washed by
methanol until the methanol was colorless. The solid was then
dissolved in CH2Cl2, and ammonia gas was bulbed through the solution
for 30 min (in order to reduce back the polymer). The color changed
from green solution (oxidized polymer) to orange (neutral polymer),
and the solution was washed with water to remove the ammonium-
iron complex salts. The solution was then washed again with 0.1 M
ammonium hydroxide solution followed by brine and dried over
magnesium sulphate. UV-vis showed no detected Fe2/Fe3. In order
to remove the short oligomers, the polymer was precipitated from the
CH2Cl2-methanol mixture.
Figure 3. Synthesis of the active NBT monomer.
iophene) (PBT) and in acceptor-substituted PBT copolymers.
The acceptor unit used here is the specially synthesized
monomer 1-(4-nitrophenyl)-2-(3-thienyl)ethene (NBT). We find
a considerable enhancement of the slow decay part of the
transient photoconductivity upon acceptor substitution, which
we attribute to charge trapping in the excited state.
In order to determine the ratio 1:m we used H-NMR measurements
of the solutions in CDCl3 on a Bruker-AC200 spectrometer. The ratio
1:m was obtained by taking the ratio of the integrated H-NMR signal
of the benzillic 2H, which belongs to the butyl substitution (chemical
shift δ ) 2.7 ppm), to that of the ortho 2H, which belongs to the
nitrobenzene substitution (δ ) 8.1 ppm). The monomer and polymer
structures were also confirmed by H-NMR measurements. The
absorption spectra of the polymers solutions in dichloromethane were
measured on an HP 8524A spectrophotometer. High-resolution mass
spectra of the monomers were taken on a Finnigan MAT-711.
For the conductivity measurements,16 thin films were mounted on
alumina (or glass) substrates by spin coating, thus obtaining films of
thickness of the order of 40-80 nm. The thickness of the films were
measured using the R-step technique, and the comparative data were
normalized to the film thickness. The dark conductivity and transient
PC were measured by employing a microswitch transmission line
configuration (Auston switch).17 Microstripline gold electrodes (width
of 0.6 mm on alumina) were deposited on the spin-coated polymer
film, leaving a gap of 0.1 mm. The back side of the substrate was
gold deposited, forming a transmission line with 50 Ω impedance with
an expected frequency response over 10 GHz. The 50 Ω impedance
matching was kept along the circuit using appropriate wiring and
connectors. For the dark conductivity measurements, a Keithly 236
source measuring unit was used for taking the I-V characteristics. An
ohmic behavior was measured up to 100 V across the 0.1 mm gap
(electric field bias of 104 V/cm). For the temperature dependence of
the dark and photoexcited conductivity measurements, the samples were
mounted onto a cold finger Air-Product LT-3-1100 cryostat in a vacuum
better than 10-4 Torr. For the transient PC measurements, one of the
electrodes was biased up to 100 V, while the other was connected to
a Tektronix 2430 digital oscilloscope via CLC144 Comlinear DC-1.1
GHz band amplifier. Excitation pulses of width of 6-8 ns at 3.5 eV
(355 nm) were obtained from the third harmonics of a Quanta-Ray
DCR-10 Nd:YAG laser, operated at a repetition rate of 10 Hz, with an
2. Experimental Section
We have selected the conjugated polymer PBT because it is highly
soluble and can be easily processed10 and substituted with HEA
(“acceptor”) monomer groups.11 The new copolymer systems contain
a random distribution of the neutral monomer 3-butylthiophene and
the active acceptor 3-HEA-thiophene. The synthesis of these copoly-
mers was carried out by an oxidative coupling procedure10,12 with the
desired concentrations of the different monomers as shown schemati-
cally in Figure 2. The HEA group is attached to the third position of
the thiophene ring in a general Wittig reaction13,14 as shown in Figure
3. We chemically prepared the homopolymer PBT and the copolymers
poly(NBT/BT) [P(NBT/BT)] with mixing ratios of 1:m (i.e., on the
average 1/m of all 3-butylthiopene (BT) monomers are substituted by
NBT).11 (The synthesis of the BT monomer is described elsewhere.15)
The synthesis of the functional monomer 3-HEA-thiophene is
described here for the monomer 1-(4-nitrophenyl)-2-(3-thienyl)ethene
(see 4, Figure 3):
3-(Bromomethyl)thiophene (2): 3-Methylthiophene (1, Figure 3)
(4 mL, 41.35 mmol) and NBS (7.36 g, 41.35 mmol) were dissolved in
100 mL CCl4. The solution was refluxed under a high-intensity lamp
for 2 h. The reaction mixture was cooled to ambient temperature and
filtrated, and the solvent was evaporated. There was no further workup
carried out. TLC showed the product 2 signature with traces of the
starting material 1.
The Phosphonium Salt of 2 (3): Triphenylphosphine (11.53 g, 44
mmol) and 2 (about 40 mmol) were dissolved in 48 mL of DMF, and
the mixture was heated up. After 5 min, the product salt 3 started to
precipitate. The mixture was then stirred for an additional 15 min,
(10) Kulszewicz-Bajer, I.; Pawlicka, A.; Plenkiewicz, J.; Pron, A.;
Lefrant, S. Synth. Met. 1989, 28, 335.
(11) Greenwald, Y.; Poplawski, J.; Wei, X.; Ehrenfreund, E.; Speiser,
S.; Vardeny, Z. V. Mol. Cryst. Liq. Cryst. 1994, 242, 145.
(12) Suginoto, R.; Takeda, S.; Yoshino, K. Chem. Express 1986, 1, 635.
(13) Campbell, R. W.; Hill, H. W., Jr. J. Org. Chem. 1973, 38, 1047.
(14) Hagen, S.; Roth, S.; Hanack, M. Synth. Met. 1991, 41-43, 1557.
(15) Tamao, K.; Kodama, S.; Nakajima, I.; Kumada, M.; Minato, A.;
Suzuki, K. Tetrahedron 1982, 38, 3347.
(16) Moses, D. Philos. Mag. 1992, B66, 1.
(17) Auston, D. H. In Picoseconds Optoelectronics DeVices; Lee, C. H.,
Ed.; Academic Press: New York, 1988; Chapter 4.
(18) Greenwald, Y.; Wei, X.; Jeglinski, S.; Poplawski, J.; Ehrenfreund,
E.; Speiser, S.; Vardeny, Z. V. Synth. Met. 1995, 69, 321.

2982 J. Am. Chem. Soc., Vol. 118, No. 12, 1996
Greenwald et al.
Figure 6. Time-resolved PC of the P(NBT/BT) 1:3 copolymer.
Figure 4. (top) Microswitch strip line (Auston Switch) configuration.
(bottom) Top view of the microstripline gold electrodes spin coated
on the polymer film. Thicknesses and sizes are given in the text.
Figure 7. Log-log plot of the PC intensity dependence of the P(NBT/
BT) 1:7 copolymer at the peak (=0 ns) and at 30 ns (the slow decay
component). The straight lines are fits to power law dependence IPC
ILR, with R = 1 for IL < 30 mW and R ) 3.4 (peak) and 4 (30 ns
delay) for 40 < IL < 90 mW.
Table 1. Activation Energy (Ea), Peak Photoconductivity, PC at
100 ns Delay, Fractional Charge for the Slow Decay (t > 10 ns)
(Qs) (see Eq 1), and Relative Bimolecular Rate Constants for the
Homopolymer PBT and Two P(NBT/BT) 1:m (m ) 3 and 7)
Copolymers Studied in the Auston Switch Configuration
Ea dark
(eV)
peak PC
PC at 100 ns
(10-5 S/cm)
(10-5 S/cm)
Qs (%) kPBT/kcop
PBT
1:7
1:3
0.68
0.74
0.78
10
9
7
0.03
0.8
2
10
70
90
1
8
21
Figure 5. Arrhenius plot of the dark conductivity: current (log scale)
Vs inverse temperature.
The time-resolved transient photocurrent decay spectrum of
a P(NBT/BT) 1:3 copolymer film at room temperature is
presented in Figure 6. A clear slow decay, which lasts up to
few microseconds, was observed. This slow decay is a strong
indication for charge trapping.
average laser power in the range of 1-100 mW, giving light intensities
of 0.1-10 mJ/(pulse cm2). The sample configuration and the circuit
involved are shown in Figure 4.
Figure 7 shows the dependence of the PC signal on the
average laser power, in the range 1-100 mW. At low
intensities, both the fast (measured at the peak at =0 delay)
and slow (30 ns delay) decay PC components are linear with
the intensity, IL, whereas at high intensities, the dependence is
superlinear, ILR, with R ) 3.4 and 4, respectively. The linear
dependence at low intensities is typical for the usual single-
photon process. The superlinear behavior at higher intensities
is indicative of multiphoton processes, which might lead to
photoionization. We stress here that all the experimental results
shown here were reversible and were obtained repeatedly while
varying the light intensity over the whole range.
3. Results
The samples’ resistivities at room temperature were on the
order of 108 Ω cm decreasing exponentially with increasing
temperature. An Arrhenius plot of the dark current is shown
in Figure 5 for PBT and two copolymers P(NBT/BT) 1:m with
m ) 7 and 3. The linear dependence reveals an activated
mechanism for the dark current flow in the polymer. Com-
parison of the Arrhenius plots of the different samples PBT and
P(NBT/BT) 1:m (m ) 3, 5, and 7) shows an increase in the
activation energy, Ea, with increasing acceptor concentration;
the values of Ea are summarized in Table 1.

Transient PhotoconductiVity of Poly(3-butylthiophene)
J. Am. Chem. Soc., Vol. 118, No. 12, 1996 2983
photocurrent. It was found that for the PBT sample Qs is only
10%. Qs increased up to 70% for the copolymer with the ratio
of 1:7 of the acceptor monomer. In the copolymer with the
ratio of 1:3, up to 90% of the charge is in the slow PC part.
The measured PC signal in the P(NBT/BT) 1:3 copolymer lasts
for few microseconds.
4. Discussion
The dark conductivity is limited by interchain charge transfer,
which is an activated process. The activation energy Ea is
inversely related to the conjugation length. Since the observed
increase in Ea in our samples is correlated with the blue shift
of the absorption peak,18 we conclude that with increasing
acceptor concentration the conjugation length decreases.
In order to account for the long time behavior of the
photocurrent, we assume that, for t > 10 ns (the system recovery
time following pulse excitation), the dominant charge recom-
bination process is bimolecular. Thus, the long time behavior
of the charge density, and of the PC current, is given by
Figure 8. Linear plot of time-resolved PC for PBT and P(NBT/BT)
1:m copolymers with m ) 3 and 7. The data plotted were taken at
room temperature and with electric field of 104 V/cm. The light intensity
is low (=2 mJ/cm2 per pulse) and corresponds to the linear part of
Figure 7.
n(t) ) (n(0)-1 + kt)-1
(2)
where n(t) is the (negative or positive) charge carrier density at
time t, n(0) ) n(t)0), and k is the apparent bimolecular rate
constant. It is important to note that this mechanism ignores
all the early processes like charge generation and monomolecular
recombination which presumably end long before the slow PC
is apparent. The bimolecular process is adequate only for those
carriers that survive the monomolecular recombination. The
solid curves in Figure 9 are the fits of the data to a bimolecular
recombination process. The parameters obtained by fitting eq
2 to the data of Figure 9 are summarized in Table 1. It is
apparent that k decreases with the acceptor concentration (Table
1). Relative to the homopolymer, it is reduced by factors of 8
and 21 for the 1:7 and 1:3 copolymers, respectively.
It is interesting to note that the bimolecular rate constants
obtained by these fits and the activation energies Ea found from
the dark conductivity measurements obey the following relation:
Figure 9. Semilog plot of the transient PC for PBT and P(NBT/BT)
1:m copolymers with m ) 3 and 7. The solid lines are fits of the
bimolecular decay (eq 2) to the data.
ln(kcop/kPBT) = -∆Ea(cop)/kBT
(3)
Comparative transient PC data of the homopolymer PBT and
P(NBT/BT) 1:m (with m ) 7 and 3) copolymers are presented
on a linear plot in Figure 8. These results were obtained at
low light intensities where the PC signal is linear with intensity.
The PBT sample shows the highest photocurrent peak value,
giving a photoconductivity of about 3 × 10-4 S/cm. It decreases
by about 10% and 30% for m ) 7 and 3, respectively. In Figure
9, we show the same data as a semilog plot in order to enhance
the small signal at long delays. For the PBT sample, the slow
component is very small and becomes nearly zero at about 100
ns (about 3 orders of magnitude smaller than the peak value).
It increases considerably as the acceptor concentration increases.
For P(NBT/BT) 1:3 copolymer, the PC current at 100 ns is
smaller than the peak value by only a factor of =10 and it is in
the range of 10-5 S/cm. This is a dramatic increase of the PC
at 100 ns in the substituted PBTsup to 2 orders of magnitude
over that of the homopolymer PBT on the same time scale.
Another measure of the relative importance of the slow decay
is the integrated fractional charge in the slow decay component,
defined as
where kcop and kPBT are the rate constants for the copolymer
and the homopolymer, respectively. ∆Ea is the difference in
the respective activation energies; i.e. ∆Ea(cop) ) Ea(cop) -
Ea(PBT). This correlation, together with the temperature
dependence of the slow part of the PC, indicates that the slow
PC arises from an activated process. Moreover, we relate the
slow component of the photocurrent decay and the dark
conductivity with the same “bottleneck” which limits the charge
transport at low charge density in the polymer. Both the dark
conductivity and the recombination process are governed by
the bimolecular or interchain charge transfer (hopping), which
is an activated process.
5. Conclusions
By investigating the slow part of the PC decay, we have
shown in this work that HEA groups, such as nitrobenzene, can
serve as traps for electrons (acceptors) in the excited state and
to delay the recombination of the photoinduced charge carriers.
The slow decay of the sample shows a bimolecular recombina-
tion process. The unique correlation between the rate constant
of the bimolecular recombination process and the activation
energy found from the dark conductivity indicates that both the
dark conductivity and the bimolecular recombination are
dominated by the same bimolecular process: the interchain
charge transfer (hopping), a temperature-activated process. The
Q ) ∞I dt ∞IPC dt
(1)
∫
∫
s
PC
τ
0
/
where τ = 12-14 ns is the system recovery time following
pulse excitation (see the abrupt change in slope from very fast
response to very slow response in Figure 8) and IPC is the

2984 J. Am. Chem. Soc., Vol. 118, No. 12, 1996
Greenwald et al.
decrease of the peak of the PC signal (at zero delay) in the
HEA-substituted copolymers is possibly the result of the bipolar
photoexcited state which lasts less than our time resolution (a
few nanoseconds). This bipolar excited state is in effect a “trap”
for the photoexcited electrons, inhibiting their recombination
with the positive charge. Thus, relative to the homopolymer,
there is a delayed charge separation in the copolymers, causing
an initial drop in the PC (peak decreases) and enhanced slow
component PC. Higher photoexcited charge concentration and
related conductivity in the nanosecond time regime may be
useful for new applications, such as electrooptic modulators,
organic photodiodes, and in photorefractive devices.
Acknowledgment. Work at the Technion was supported by
the Israel Science Foundation administered by the Israel
Academy of Sciences and Humanities (335/94). Work at the
Hebrew University was supported by the VW Foundation and
The Ministry of Science and the Arts, Jerusalem, Israel. We
would like to thank Dr. D. Moses from the University of
California, Santa Barbara, for his stimulating discussions and
help and Mr. Haim Chayat from the Hebrew University for
useful comments.
JA953762G