Benzodithiophene and imide-based copolymers for photovoltaic applications

Article abstract of DOI:10.1021/cm2038427

Conjugated alternating copolymers were designed with low optical band gaps for organic photovoltaic (OPV) applications by considering quinoid resonance stabilization. Copolymers of thienoisoindoledione (TID) and benzodithiophene (BDT) had appreciably lower band gaps (by ～0.4 eV) than copolymers of thienopyrroledione (TPD) and BDT. In addition to intramolecular charge transfer stabilization (i.e., the "push-pull" effect), the former copolymer's quinoid resonance structure is stabilized by a gain in aromatic resonance energy in the isoindole unit. Additionally, the HOMO levels of the copolymers could be tuned with chemical modifications to the BDT monomer, resulting in open circuit voltages of greater than 1 V in photovoltaic devices. Despite the optimized band gap, TID containing polymers displayed lower photoconductance, as determined by time-resolved microwave conductivity, and decreased device efficiency (2.1% vs 4.8%) as compared with TPD analogues. These results were partially attributed to morphology, as computational modeling suggests TID copolymers have a twisted backbone, and X-ray diffraction data indicate the polymer films do not form ordered domains, whereas TPD copolymers are considerably more planar and are shown to form partially ordered domains.

Full text of DOI:10.1021/cm2038427

Article
pubs.acs.org/cm
Benzodithiophene and Imide-Based Copolymers for Photovoltaic
Applications
Wade A. Braunecker,*,† Zbyslaw R. Owczarczyk,† Andres Garcia,† Nikos Kopidakis,† Ross E. Larsen,†
Scott R. Hammond,† David S. Ginley,† and Dana C. Olson†
†National Renewable Energy Laboratory, 1617 Cole Boulevard, Golden, Colorado 80401, United States
S
* Supporting Information
ABSTRACT: Conjugated alternating copolymers were designed with low optical band gaps
for organic photovoltaic (OPV) applications by considering quinoid resonance stabilization.
Copolymers of thienoisoindoledione (TID) and benzodithiophene (BDT) had appreciably
lower band gaps (by ∼0.4 eV) than copolymers of thienopyrroledione (TPD) and BDT. In
addition to intramolecular charge transfer stabilization (i.e., the “push-pull” effect), the former
copolymer’s quinoid resonance structure is stabilized by a gain in aromatic resonance energy
in the isoindole unit. Additionally, the HOMO levels of the copolymers could be tuned with
chemical modifications to the BDT monomer, resulting in open circuit voltages of greater than 1 V in photovoltaic devices.
Despite the optimized band gap, TID containing polymers displayed lower photoconductance, as determined by time-resolved
microwave conductivity, and decreased device efficiency (2.1% vs 4.8%) as compared with TPD analogues. These results were
partially attributed to morphology, as computational modeling suggests TID copolymers have a twisted backbone, and X-ray
diffraction data indicate the polymer films do not form ordered domains, whereas TPD copolymers are considerably more planar
and are shown to form partially ordered domains.
KEYWORDS: pyrroledione, isoindoledione, benzodithiophene, solar cell, low band gap polymer
efficient at stabilizing this delocalization is the use of donor−
acceptor alternating copolymers. Intramolecular charge transfer
between alternating units of electron-donating and electron-
accepting moieties effectively promotes conjugation throughout
the polymer backbone by quinoid resonance stabilization,
thereby appreciably narrowing the band gap through the so-
called “push-pull” effect.13,19 Another approach to achieve this
stabilization employs fused aromatic units, as in polyisothia-
naphthene (PITN), wherein the dearomatization of the
thiophene ring to assume a quinoid structure is accompanied
by a gain in aromatic resonance energy in the fused benzene
ring, resulting in a band gap as much as one full electronvolt
lower than in polythiophene.20 Since the discovery of PITN,
several other reports have appeared on this class of materials,
describing polymers with structural variations of the
isothianaphthene unit,21,22 as well as polymers based on
thienopyrazine23 and thienothiophene.24 Figure 1 illustrates a
new copolymer investigated in this report for which the quinoid
structure is stabilized both by intramolecular charge transfer as
well as aromatic resonance.
INTRODUCTION
■
The ongoing development of polymer solar cells as low cost,
lightweight, flexible alternatives to silicon-based inorganic
devices is advancing the field of organic photovoltaics (OPV)
and helping address the ever-increasing global energy
demand.1−3 The most successful OPV devices to date are
based on the bulk heterojunction (BHJ) concept, which
typically have an interpenetrating network of photoactive
electron-donating polymers and fullerene-based acceptors.4−6
Indeed, the field has witnessed explosive growth and develop-
ment in just the past few years, with rational design efforts
playing a key role in fine-tuning the electronic properties of the
polymer donors that in turn enhance power conversion
efficiencies (PCEs).7−10 Low band gap polymers, whose
absorption better overlaps with the solar spectrum, are now
being targeted (thereby maximizing photon absorption and the
short-circuit current, JSC), in conjunction with polymers having
lower highest occupied molecular orbital (HOMO) energy
levels (thereby maximizing the open-circuit voltage, VOC).11−13
While certain parameters that clearly and dramatically affect
device performance are more difficult to predict and control,
including polymer morphology14 and polymer-fullerene inter-
calation,15 rational design efforts are beginning to fuel clear
progress in the field. Reports of device efficiencies now
approaching 10%16 confirm that OPV is becoming a viable
technology.
Recently, the photovoltaic performances of alternating
copolymers containing benzodithiophene (BDT) as the
donating moiety have attracted much attention.9,25,26 Among
these systems, polymers containing derivatives of thieno[3,4-
b]thiophene-2-carboxylates (TT) as the electron-accepting unit
have shown great promise, with PCEs as high as 7.4%.27
Designing a semiconducting polymer with a narrow band gap
requires consideration of the conjugated polymer’s quinoid
resonance structures or delocalization along its backbone (see
Figure 1).17,18 One strategy that has proven particularly
Received: December 23, 2011
Revised: March 5, 2012
Published: March 6, 2012
© 2012 American Chemical Society
1346
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
Figure 1. Thienoisoindoledione-benzodithiophene alternating copolymer whose quinoid structures can be stabilized through intramolecular charge
transfer and aromatic resonance.
Scheme 1. Synthesis of Monomer Derivatives of TPD (2), TID (7), and BDT (9, 14, 15)
Interestingly, these TT units stabilize the quinoid resonance
structure both through intramolecular charge transfer as well as
aromatic resonance. This stabilization is ultimately responsible
for the narrow polymer band gap (∼1.6 eV). However, these
polymers generally possess a relatively shallow HOMO energy
level (typically around −5.0 eV). As the VOC of these BHJ
devices are directly related to the energy difference between the
HOMO level of the donor polymer and the LUMO level of the
fullerene acceptor,11 the shallow HOMO contributes to a
relatively low VOC. Another successful copolymer of BDT with
deeper HOMO levels (but a wider band gap) than that with
TT, which was independently reported by three groups,28−30
employs 5-alkyl-thieno[3,4-c]pyrrole-4,6-dione (TPD) with an
alkoxy-chain derivative of BDT, 4,8-di(2-ethylhexyloxyl)benzo-
[1,2-b:4,5-b′]dithiophene. The optical band gap for these
1347
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
3,4-Bis(hydroxymethyl)-thiophene (3). A slurry of thiophene-
3,4-dicarboxylic acid (10.0 g, 58.0 mmol) in 100 mL of anhydrous 1,2-
dichloroethane was first converted to the acid chloride by following
the procedure described above for compound (2) using oxalyl
chloride. After the conversion and removal of solvent, 50 mL of dry
THF was added. In a separate flask, 2.0 eq. of lithium aluminum
hydride (4.40 g, 116 mmol) was added to 500 mL of dry THF and
stirred in an ice bath. The acid chloride solution was transferred
dropwise via cannula to the LAH slurry. The mixture was stirred at r.t.
for three additional hrs. The THF was evaporated, and a saturated
solution (100 mL) of NaCl was then added dropwise. The solution
was acidified (pH ∼ 2) and then washed 5× with 200 mL of ether.
The combined organic extracts were washed with 20 mL of water and
then dried over MgSO4. The ether was removed to give 7.1 g (85%) of
systems was reported as ∼1.8 eV, with a HOMO approximately
−5.4 eV, and optimized efficiencies varying from 4.2% to 6.8%.
The reported band gaps and HOMO levels for these TPD-
BDT systems suggest room for further optimization. We report
herein on our efforts to further narrow the band gap of this
system by employing an electron-accepting moiety of similar
strength to TPD, but whose quinoid structure can also be
stabilized by aromatic resonance, 5-octyl-thieno[3,4-f ]-
isoindole-5,7-dione (TID). This monomer has been used to
synthesize polymers with band gaps as low as 1.10 eV for
certain electrochromic devices22,31 but, to our knowledge, has
not been used specifically for any OPV applications. Addition-
ally, we attempt to push the HOMOs of our donor polymers a
little deeper by employing alkyl-chain derivatives of BDT
(Scheme 1). Literature reports have demonstrated this strategy
can improve the VOC of similar copolymers by more than 100
mV as compared to alkoxy-chain derivatives of BDT.9
Note, this synthetic effort is not a serendipitous approach to
combining known donor monomers with known acceptors in
different combinations to achieve “new” OPV polymers. Rather,
we primarily seek to contribute to the ongoing design effort
aiming to fine-tune band gaps and HOMO levels of OPV
polymers. Some additional results for the two basic polymer
structures (TID vs TPD containing polymers) are provided to
help understand their performance in solar cell devices,
including the photoconductance of polymer films, as
determined by time-resolved microwave conductivity; the
presence or absence of ordered domains in the films, as
determined by X-ray diffraction; and the relative planarity of
the polymer backbones, as determined by computational
modeling.
1
the title compound as a yellow oil. H NMR (400 MHz, CDCl3): δ
7.10 (s, 2H), 4.49 (s, 4H), 3.75 (br, 2H). 13C NMR (100 MHz,
CDCl3): δ 140.7, 125.1, 59.5.
Thiophene-3,4-dicarbaldehyde (4). Compound (3) (6.50 g,
45.1 mmol) was stirred in 400 mL of anhydrous methylene chloride at
r.t. for 3 h with 3.0 eq. of pyridinium chlorochromate (29.2 g, 135
mmol) and 40 g of crushed molecular sieves. The reaction mixture was
then filtered through a pad of silica gel and washed with an additional
1 L of methylene chloride. The solvent was evaporated to give 4.42 g
(70%) of the title compound as an off-white solid, mp 76 °C. 1H NMR
(400 MHz, CDCl3): δ 10.28 (s, 2H), 8.20 (s, 2H). 13C NMR (100
MHz, CDCl3): δ 186.1, 140.5, 137.9.
1-Octyl-pyrrole-2,5-dione (5). This compound was synthesized
according to a literature procedure34 for N-alkyl-pyrrole-2,5-dione
derivatives in 85% yield as a light tan solid, mp 36 °C. 1H NMR (400
MHz, CDCl3): δ 6.65 (s, 2H), 3.48 (m, 2H), 1.58 (br m, 2H), 1.27 (br
m, 10H), 0.87 (t, 3H, J = 7.0 Hz). 13C NMR (100 MHz, CDCl3): δ
171.1, 134.2, 38.1, 32.0, 29.3, 29.3, 28.7, 26.9, 22.8, 14.3.
6-Octyl-thieno[3,4-f ]isoindole-5−7-dione (6). This compound
was synthesized based on a literature procedure35 for the elongation of
acene derivatives. Dialdehyde (4) (1.50 g, 10.7 mmol) and maleimide
(5) (2.07 g, 10.7 mmol) were dissolved in 100 mL of anhydrous
methylene chloride and stirred in an ice bath. To this cooled solution,
a 10 mL solution of methylene chloride containing 0.1 eq. DBU (160
μL, 1.07 mmol) and 1.5 eq. tri-n-butylphosphine (4.00 mL, 16.1
mmol) was added dropwise. The reaction was stirred for 1 h at r.t.,
after which it was concentrated in vacuo and the residue was purified
via flash chromatography in hexane:ethyl acetate (4:1). After removing
the solvent, the red solid was recrystallized twice from methanol to
give 2.26 g (67%) of off-white crystals. 1H NMR (400 MHz, CDCl3):
δ 8.13 (s, 2H), 8.02 (s, 2H), δ 3.70 (t, 2H, J = 7.4 Hz), 1.68 (m, 2H),
1.38−1.2 (m, 10H), 0.85 (t, 3H, J = 7.0 Hz). 13C NMR (100 MHz,
CDCl3): δ 168.2, 138.4, 126.0, 122.8, 119.5, 38.9, 32.0, 29.4, 28.6, 27.1,
22.8, 14.3.
EXPERIMENTAL SECTION
■
General. All reagents employed in this study were obtained from
commercial sources at the highest available purity and used without
further purification, unless otherwise noted. All reactions were
performed under dry N2. Methylene chloride, toluene, and THF
were purified by passing through alumina in an MBraun solvent
purification system. Column chromatography was performed with
Fluka Silica Gel 60 (220−440 mesh). All small molecules were
characterized by 1H NMR (400 MHz) and 13C NMR (100 MHz) on a
Varian Unity Inova. Monomers were >99% pure as determined by 1H
NMR. UV−vis absorption measurements were performed using a
Hewlett-Packard 8453 UV−vis spectrophotometer.
1,3-Dibromo-6-octyl-thieno[3,4-f ]isoindole-5−7-dione (7).
This compound was synthesized according to a modified literature
procedure.36 NBS (2.05 g, 11.5 mmol) was added to a stirred solution
of compound (6) (1.77 g, 5.61 mmol) in a mixture of 75 mL of
chloroform and 75 mL of acetic acid. The mixture was stirred at r.t.
overnight, and then stirred at 60 °C for 2 h. The mixture was
quenched with water, the organic layer was separated, and the water
layer was extracted 3× with 10 mL of chloroform. The combined
chloroform extracts were washed with a sodium bicarbonate solution.
After drying over MgSO4, the solvent was evaporated, and the residue
was purified via flash chromatography using methylene chloride:hex-
ane (2:1) as the eluent to afford 2.0 g (76%) of a bright yellow solid.
1H NMR (400 MHz, CDCl3): δ 7.98 (s, 2H), δ 3.70 (t, 2H, J = 7.4
Hz), 1.68 (m, 2H), 1.38−1.2 (m, 10H), 0.85 (t, 3H, J = 7.0 Hz). 13C
NMR (100 MHz, CDCl3): δ 167.2, 138.1, 127.5, 118.6, 110.1, 38.9,
32.0, 29.4, 28.6, 27.1, 22.8, 14.3.
Synthesis. 1,3-Dibromo-5-octyl-4H-thieno[3,4-c]pyrrole-
4,6(5H)-dione (2). This compound was synthesized according to a
modified literature procedure.32 A slurry of 2,5-dibromothiophene-3,4-
dicarboxylic acid (1)33 (2.25 g, 6.86 mmol) was stirred in 30 mL of
anhydrous 1,2-dichloroethane. One drop of DMF was added, followed
by the dropwise addition of oxalyl chloride (1.83 mL, 20.6 mmol).
After 20 min at r.t. and much vigorous bubbling, the solution became
homogeneous. It was then heated at 60 °C for one hour, after which
the solvent was removed by boiling off under N2. The crude sample
was dried under vacuum to give ∼2.5 g of the acid chloride, which was
used without further purification. 1.0 eq. of N-octylamine (1.14 mL,
6.86 mmol) was added dropwise at r.t. to the solid acid chloride under
N2 (fuming HCl evolves), after which the reaction flask was heated to
130 °C for 30 min. After cooling to r.t., the sticky solid was dissolved
in 20 mL of ethyl acetate, washed with a saturated solution of sodium
bicarbonate, and dried over MgSO4. The product was then purified by
flash chromatography on silica gel with hexane:ethyl acetate (10:1).
After removal of the solvent, the solid was twice recrystallized from
4,8-Bis(2-ethylhexyloxy)benzo[1,2-b:4,5-b′]dithiophene (8).
The precursor, benzo[1,2-b:4,5-b′]dithiophene, was synthesized
according to a literature procedure.37 Compound (8) was obtained
according to a modified literature procedure9 by mixing benzo[1,2-
b:4,5-b′]dithiophene (1.0 g, 4.5 mmol), Na2S2O4 (1.74 g, 9.97 mmol),
and tetrabutylammonium bromide (1.61 g, 5.0 mmol) in 50 mL of
1
MeOH to afford 1.7 g (59%) of the title compound, mp 104 °C. H
NMR (400 MHz, CDCl3): δ 3.59 (t, 2H, J = 7.4 Hz), 1.62 (m, 2H),
1.25 (m, 10H), 0.87 (t, 3H, J = 7.0 Hz). 13C NMR (100 MHz,
CDCl3): δ 160.6, 135.0, 113.1, 39.1, 32.0, 29.3, 28.5, 27.0, 22.8, 14.3.
1348
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
water for 10 min. 80 mL of THF was then added, along with 10 g of
NaOH. The reaction was stirred for 2 h at r.t. while purging with N2.
2-Ethylhexyl iodide (1.95 mL, 10.9 mmol) was added, and the reaction
was stirred at 50 °C overnight. The solution was diluted with 100 mL
of water and extracted with 100 mL of ethyl acetate. The organic
extract was dried over MgSO4 and the solvent was evaporated.
Column chromatography using methylene chloride:hexane (1:2) as
the eluent yielded 0.72 g (36%) of the title compound as a light yellow
oil. 1H NMR (400 MHz, CDCl3): δ 7.48 (d, 2H, J = 5.6 Hz), 7.36 (d,
2H, J = 5.6 Hz), 4.19 (d, 4H, J = 5.6 Hz), 1.8 (m, 2H), 1.75−1.25 (m,
16H), 1.02 (t, 6H, J = 7.2 Hz), 0.94 (t, 6H, J = 7.2 Hz). 13C NMR
(100 MHz, CDCl3): δ 144.9, 131.7, 130.2, 126.1, 120.5, 76.3, 40.9,
30.7, 29.5, 24.1, 23.4, 14.4, 11.5.
2,6-Bis(trimethyltin)-4,8-bis(2-ethylhexyloxy)benzo[1,2-
b:4,5-b′] dithiophene (9). Compound (9) was obtained according
to a literature procedure.9 1H NMR (400 MHz, CDCl3): δ 7.51 (s,
2H), 4.18 (d, 4H, J = 5.6 Hz), 1.90−1.35 (m, 18H), 1.02 (t, 6H, J =
7.2 Hz), 0.96 (t, 6H, J = 7.2 Hz). 13C NMR (100 MHz, CDCl3): δ
143.6, 140.7, 134.2, 133.2, 128.3, 75.9, 41.0, 30.9, 29.6, 24.2, 23.5, 14.5,
11.7, −8.1.
4,8-Bis(3-ethylhept-1-ynyl)benzo[1,2-b:4,5-b′]dithiophene
(10). 3-Ethylhept-1-yne was synthesized according to a literature
procedure,38 and it was used to synthesize compound (10) according
to a modified literature procedure.9 Isopropylmagnesium chloride
(2M, 1.84 mL, 3.68 mmol) was added dropwise to a solution of 3-
ethylhept-1-yne (0.5 g, 4.03 mmol) at r.t. in 25 mL of THF. The
reaction mixture was stirred at 60 °C for 2 h. After cooling to r.t.,
benzo[1,2-b:4,5-b′]dithiophene (0.385 g, 1.75 mmol) was added, and
the mixture was stirred at 60 °C for an additional 2 h. After cooling to
r.t., 2 g of SnCl2 in 4 mL of a 10% HCl solution was added dropwise.
The reaction was stirred for 1 more h and then poured into 100 mL of
water. It was extracted twice with 50 mL of hexane. The combined
organic phase was dried over MgSO4 and the solvent was evaporated.
The residue was purified using column chromatography on silica with
hexane:methylene chloride (3:1) as the eluent to yield 0.550 g (72%)
of the title compound. 1H NMR (400 MHz, CDCl3): δ 7.57 (d, 2H, J
= 5.2 Hz), 7.50 (d, 2H, J = 5.6 Hz), 2.70 (m, 2H), 1.75−1.50 (m,
12H), 1.43 (m, 4H), 1.19 (t, 6H, J = 7.4 Hz), 0.97 (t, 6H, J = 7.4 Hz).
13C NMR (100 MHz, CDCl3): δ 140.5, 138.4, 127.8, 123.4, 112.4,
mmol) was dissolved in 20 mL of dry THF and cooled to −78 °C
under N2. A butyllithium solution (2.5 M, 0.594 mL, 1.48 mmol) was
added dropwise, and the reaction mixture was stirred for 30 min at
−78 °C, warmed to r.t., and then cooled back down to −78 °C. A
trimethyltin chloride solution (1 M, 1.7 mL, 1.70 mmol) was added
dropwise at −78 °C. The reaction was warmed to r.t. and stirred
overnight. The mixture was quenched with 50 mL of water and
extracted twice with 50 mL of hexanes. The combined organic phase
was dried over MgSO4 and the solvent was evaporated. The residue
was twice recrystallized from isopropanol to yield 0.410 g (80%) of the
title compound. 1H NMR (400 MHz, CDCl3): δ 7.48 (s, 2H), 3.15 (t,
4H, 8.0 Hz). 1.74 (m, 4H), 1.50−1.30 (m, 18H), 0.95 (m, 12H), 0.44
(s, 18H). 13C NMR (100 MHz, CDCl3): δ 141.6, 140.4, 136.9, 129.9,
128.0, 39.6, 33.0, 32.9, 30.9, 29.2, 26.2, 23.4, 14.5, 11.3, −8.2.
2,6-Bis(trimethyltin)-4,8-(4-butylphenethyl)benzo[1,2-b:4,5-
b′]dithiophene (15). Compound (15) was synthesized in an
analogous manner as compound (14). 1H NMR (400 MHz,
CDCl3): δ 7.46 (s, 2H), δ 7.24 (d, 2H, J = 8.0 Hz), 7.14 (d, 4H, J
= 8.0 Hz), 3.50 (m, 4H), 3.07 (m, 4H), 2.61 (t, 4H, J = 7.8 Hz), 1.60
(m, 4H), 1.37 (sextet, 4H, J = 7.4 Hz), 0.93 (t, 6H, J = 7.4 Hz), 0.45
(s, 18H). 13C NMR (100 MHz, CDCl3): δ 141.5, 140.6, 140.5, 139.2,
136.9, 129.6, 128.5, 128.3, 126.7, 35.75, 35.3, 33.8, 22.4, 14.0, −8.3.
Typical Procedure for Polymer Synthesis. Compound 2 (88.0
mg, 0.21 mmol), compound 9 (161 mg, 0.21 mmol), Pd2(dba)3 (2
mol %), and tri(o-tolyl)phosphine (8 mol %) were placed in a flask,
purged with three nitrogen/vacuum cycles, and subsequently dissolved
in 5 mL of dry oxygen free chlorobenzene. The mixture was stirred for
36 h at 110 °C, after which 20 μL of 2-bromothiophene was injected
as a capping agent. The reaction was stirred for 2 h at 110 °C before
20 μL of 2-(tributyltin)thiophene was injected to complete the end-
capping. After an additional 2 h of stirring, a complexing ligand (N,N-
diethylphenylazothioformamide)39 was stirred with the polymer to
remove any residual catalyst before being cooled to r.t. and
precipitated into methanol (100 mL). The precipitate was purified
via Soxhlet extraction overnight with methanol, for 2 h with acetone,
and finally was collected with chloroform. The chloroform solution
was then concentrated by evaporation and precipitated into methanol
(200 mL) and filtered off as a dark purple solid. Typical yields were
around 80%.
Polymer Molecular Weight Determination. Polymer samples
were dissolved in HPLC grade chloroform (∼1 mg/mL), stirred and
heated at 50 °C for several hours, stirred overnight at r.t., and then
filtered through a 0.45 μm PTFE filter. Size exclusion chromatography
was then performed on a PL-Gel 300 × 7.5 mm (5 μm) mixed D
column using an Agilent 1200 series autosampler, inline degasser, and
refractometer. The column and detector temperatures were 35 °C.
HPLC grade chloroform was used as eluent (1 mL/min). Linear
polystyrene standards were used for calibration. The same general
procedure was performed for larger scale preparatory GPC work. 4.5
mL of a ∼3 mg/mL polymer solution in HPLC grade chloroform was
injected on two PL-Gel 300 × 25 mm (10 μm) mixed D columns
connected in series. An Agilent 1200 series autosampler, inline
degasser, and diode array detector were employed. The column and
detector temperatures were 25 °C. HPLC grade chloroform was used
as eluent (10 mL/min).
Cyclic Voltammetry. All voltammograms were recorded at 25 °C
with a CH Instruments Model 600D potentiostat. Unless otherwise
specified, measurements were carried out under nitrogen at a scanning
rate of 0.1 V s−1 using a platinum wire as the working electrode and a
platinum wire as the counter electrode. Potentials were measured vs
Ag/Ag+ (and calibrated vs Fc/Fc+) using 0.01 M AgClO4 and a 0.1 M
Bu4NBF4 salt bridge to minimize contamination of the analyte with
Ag+ ions. Polymer films were drop cast onto a platinum wire working
electrode from a 1 mg/mL chloroform solution and dried under a
stream of nitrogen prior to measurement in a 0.1 M Bu4NBF4
acetonitrile solution.
103.9, 78.6, 34.9, 30.0, 28.5, 22.9, 20.9, 14.4, 12.3.
4,8-Bis((4-butylphenyl)ethynyl)benzo[1,2-b:4,5-b′]-
dithiophene (11). Compound (11) was synthesized in an analogous
manner as compound (10) using commercially available 1-butyl-4-
1
ethynlbenzene. H NMR (400 MHz, CDCl3): δ 7.70 (d, 2H, J = 5.2
Hz), 7.60 (d, 4H, J = 8.0 Hz), δ 7.57 (d, 2H, J = 5.6 Hz), δ 7.22 (d,
4H, J = 8.0 Hz), 2.65 (t, 4H, J = 7.8 Hz), 1.61 (m, 4H), 1.38 (sextet,
4H, J = 7.4 Hz), 0.94 (t, 6H, J = 7.4 Hz). 13C NMR (100 MHz,
CDCl3): δ 144.3, 140.5, 138.4, 131.9, 128.9, 128.2, 123.5, 120.3, 112.3,
99.7, 85.4, 35.9, 33.6, 22.6, 14.2.
4,8-Bis(3-ethylheptyl)benzo[1,2-b:4,5-b′]dithiophene (12).
Compound (12) was synthesized according to a modified literature
procedure.9 To a solution of 10 (0.55 g, 1.26 mmol) in 20 mL of
anhydrous THF was added 0.120 g of 10% Pd/C, and the mixture was
stirred vigorously for 20 h at r.t. under 30 atm of H2. The mixture was
then filtered through a Celite pad to remove the Pd/C, and the residue
was purified by column chromatography on silica with hexane as the
eluent to yield 0.300 g (54%) of the title compound as a white solid.
1H NMR (400 MHz, CDCl3): δ 7.47 (d, 2H, J = 7.6 Hz), 7.46 (d, 2H,
J = 7.6 Hz), 3.15 (m, 4H). 1.75 (m, 4H), 1.55−1.20 (m, 18H), 0.95
(m, 12H). 13C NMR (100 MHz, CDCl3): δ 137.4, 136.0, 129.5, 126.1,
121.9, 39.6, 33.1, 33.0, 31.0, 29.2, 26.1, 23.4, 14.4, 11.2.
4,8-Bis(4-butylphenethyl)benzo[1,2-b:4,5-b′]dithiophene
(13). Compound (13) was synthesized in an analogous manner as
compound (12). 1H NMR (400 MHz, CDCl3): δ 7.45 (d, 2H, J = 7.6
Hz), δ 7.43 (d, 2H, J = 7.6 Hz), 7.20 (d, 4H, J = 8.0 Hz), 7.13 (d, 4H,
J = 8.0 Hz), 3.46 (m, 4H), 3.05 (m, 4H), 2.60 (t, 4H, J = 7.8 Hz),
1.60 (m, 4H), 1.37 (sextet, 4H, J = 7.4 Hz), 0.93 (t, 6H, J = 7.4 Hz).
2,6-Bis(trimethyltin)-4,8-(3-ethylheptyl)benzo[1,2-b:4,5-b′]-
dithiophene (14). Compound (14) was synthesized according to a
modified literature procedure.9 Compound (12) (0.300 g, 0.675
Theory. Density functional theory (DFT) and time-dependent
density functional theory (TDDFT) were used to predict the
properties of the polymers reported in this work for hydrogen-
terminated oligomers with n = 1,2,3,4. All calculations were performed
1349
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
with the default settings in the Gaussian 09 electronic structure
package, revision B.01.40 The geometric structure of each oligomer was
optimized in vacuum using the Becke-style three-parameter density
functional with the Lee−Yang−Parr correlation function (B3LYP)
with the 6-31G(d) basis set; subsequently, diffuse functions were
included (6-31+G(d)) for calculation of the orbital energies and
optical absorption spectra of the optimized structures as described
below. Including diffuse functions had little effect on the predicted
absorption spectra; the main change was a systematic shift of the
molecular orbital energies down by ∼200−300 meV. For a few
structures, calculations with larger basis sets were run, including
additional polarization functions (6-31+G(d,p)). Little change (<20
meV) in the orbital and oligomer energies was found in going from the
6-31+G(d) basis to the 6-31+G(d,p) basis, so only the former results
are reported here. We also calculated the energy associated with
structural relaxation following removal of an electron, with a 6-31G(d)
basis, by taking the difference in energy between an oligomer cation at
the optimized geometry for the neutral structure and a geometrically
optimized oligomer cation, in vacuum.
Device Fabrication. Patterned ITO/glass substrates (Thin Films
Devices, Inc.) for photovoltaic measurements were cleaned prior to
device fabrication by sonication in acetone and isopropanol followed
by an O2 plasma treatment at 800 m torr O2 pressure and 150 W
power for 5 min. Devices with the standard architecture ITO/
PEDOT:PSS/BHJ/Ca/Al, where PEDOT:PSS is poly(3,4-
ethylenedioxythiophene):poly(styrenesulfonate), were fabricated by
spin coating PEDOT:PSS (Clevios P VP Al 4083), subsequently
annealing at 140 °C for 30 min, followed by spin coating of the
corresponding polymer:PC61BM solution, and completed by thermal
deposition of 20 nm of Ca and 100 nm of Al electrodes with an area of
11 mm2 via a shadow mask at a base pressure of ∼3 × 10−8 Torr. The
blend ratio of polymer to PC61BM was altered from 1:1 to 1:4 (not
shown), and the optimal conditions found were used for these studies.
P1 BHJ films with ratio 1:2 P1:PC61BM were deposited from either 10
mg/mL polymer concentration o-dichlorobenzene (DCB) solutions or
8 mg/mL polymer concentration chlorobenzene (CB) solutions
containing 4% by volume o-diiodooctane cosolvent. P2 BHJ films were
also deposited from 1:2 P2:PC61BM 10 mg/mL polymer concen-
tration DCB solutions. For P4 BHJ films 1:1 P4:PC61BM 7 mg/mL
polymer concentration CB solutions containing 4% by volume
diiodoocatne cosolvent were spin coated. All BHJ films were fabricated
by spin coating filtered 70 °C solutions with 0.45 μm PTFE filters.
Current−voltage measurements of PV devices under a 100 mW/cm2
illumination intensity (supplied by a tungsten halogen lamp and
monitored with Hamamatsu Si photodiodes equipped with a KG5
filters) were carried out with a Keithley 236 source-measuring unit in a
N2 atmosphere. The device mismatch factor (1.05 for P1 and P2 and
1.14 for P4) was measured and calculated according to a literature
reference41 and utilized to adjust the light J − V curves.
quantum efficiency of free carrier generation per photon absorbed, and
Σμ is the sum of the mobilities of electrons and holes.43
X-ray Diffraction. X-ray diffraction measurements were performed
with a Rigaku D/MAX-2500H goniometer (185 mm Bragg−Brentano
geometry) equipped with a rotating anode X-ray generator (40 kV,
200 mA) using Cu Kα radiation with a wavelength of 0.154 nm and
detected with a scintillation counter filtered with a single-bounce
graphite monochromator. The scan rate was 1° min−1 in the 2θ range
of 2° to 10° using 0.02° steps. Analyses were conducted on polymer
films prepared on quartz substrates under the same conditions as their
respective optimized devices.
RESULTS AND DISCUSSION
■
Monomer Synthesis. Synthetic routes for the five
comonomers employed in this study are depicted in Scheme
1. The brominated TPD monomer (compound 2) was
synthesized according to a literature procedure32 in three
steps starting from commercially available thiophene-3,4-
dicarboxylic acid. An eight step synthetic procedure for the
brominated derivative of TID (compound 7) has been
described in the literature,22,36 but we present an improved
synthetic procedure for this compound based on an elongation
protocol for acene derivatives.35 Thiophene-3,4-dicarbaldehyde
(4) was synthesized in two steps from the carboxylic acid
starting material (LiBH4 reduction, then oxidation with
pyridinium chlorochromate). The three ring TID heterocycle
6 was produced from a double condensation type reaction
between 4 and N-octylmaleimide (5) that proceeded smoothly
when catalyzed by tributylphosphine. The desired TID
monomer 7 was then obtained according to a literature
bromination procedure36 with N-bromosuccinimide. The
monomer derivatives of BDT (9, 14, and 15) were synthesized
according to modified literature procedures9 described in the
Experimental Section.
Polymer Synthesis. The six copolymers in this study
(illustrated in Figure 2) were synthesized by a palladium-
Time-Resolved Microwave Conductivity. TRMC is a pump−
probe technique that can be used to measure the photoconductance of
a film without the need for charge collection at electrical contacts.42,43
The details of the experimental methodology have been presented
elsewhere.43,44 In brief, the sample is placed in a microwave cavity at
the end of an X-band waveguide operating at ca. 9 GHz and is
photoexcited through a grid with a 5 ns laser pulse from an OPO
pumped by a Nd:YAG laser. The relative change of the microwave
power, P, in the cavity, caused by the photoinduced generation of free
electrons and holes, is related to the transient photoconductance, ΔG,
by ΔP/P = −KΔG, where the calibration factor K is experimentally
determined individually for each sample. Taking into account that the
electrons and holes are generated in pairs, the photoconductance can
be expressed as43
Figure 2. Polymers P1-P6 synthesized through palladium-catalyzed
Stille coupling.
catalyzed Stille coupling at 110 °C over 36 h in chlorobenzene.
After this time, the polymers were end-capped with thiophene
reagents, as it has been demonstrated that such end-capping
can improve the performance of photovoltaic devices.45
A
palladium scavenger39 was then stirred with the polymer to
complex the catalyst before the polymer was precipitated into
MeOH, after which the polymer was purified via Soxhlet
extraction for 12 h with MeOH and 2 h with acetone. To
ensure the majority of any oligomeric material was removed
ΔG = βq F I (ϕ· μ)
∑
A 0
(1)
e
where qe is the elementary charge, β = 2.2 is the geometric factor for
the X-band waveguide used, I0 is the incident photon flux, FA is the
fraction of light absorbed at the excitation wavelength, ϕ is the
1350
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
from the polymer sample, the polymers were all purified with
preparatory gel permeation chromotagraphy (GPC), 15 mg at a
time. All of the polymers were soluble in chloroform at r.t., with
the exception of P3, whose extreme insolubility, even in boiling
o-dichlorobenzene, rendered its continued study impractical.
Molecular weight data for the remaining purified samples is
presented in Table 1.
attain polymer samples with a systematic range of molecular
weights. Such a tool can help identify the “sweet spot” in
molecular weights that give optimal solar cell device perform-
ance, allowing the investigator to then fine-tune polymerization
conditions to attain those weights on a larger scale. Such a
study is currently underway in our lab for the polymers in this
manuscript and will be presented elsewhere.
Optical Characterization of Polymers. The absorption
spectra of the polymers have been measured both in solution
(chloroform, r.t.) and in the solid state as thin films. Three
important observations are worth noting. First, the optical band
gaps of the polymers vary as a function of the electron-donating
strength of the substituents on the BDT unit. The absorption
onsets (recorded where the slope of the absoption edge crossed
the baseline) for polymers with alkoxy substituents on BDT
(P1 and P4) are red-shifted ∼25−40 nm as compared to their
respective alkyl analogues P2 and P5, a function of the
increased electron-donating strength of the alkoxy vs alkyl
groups (and hence, increased charge transfer and quinoid
stabilization). Second, the absorption onsets of TID containing
polymers (P4, P5) are dramatically red-shifted (by ∼200 nm in
the solid state) as compared to their TPD analogs (P1, P2), a
reflection of the enhanced stabilization of the quinoid structure
in the former polymers due to aromatic resonance stabilization.
Finally, the difference in a given polymer’s solution and film
absorption onset is dramatically larger for the TID containing
polymers than those containing TPD. The solid-state onsets of
P1 and P2 are red-shifted 15 nm or less compared to their
solution spectra, while onsets of P4−P6 red-shift up to 100 nm
in the solid state (Figure 4). The latter observation may
possibly be ascribed to the propensity of the TID containing
polymers to form more planar structures in the solid state than
they do in solution (see also the computational modeling
results below).
Electrochemical Characterization of Polymers. Cyclic
voltammetry was utilized to evaluate the HOMO energy levels
of polymers P1, P2, and P4−P6. The data are summarized in
Table 1 and illustrated in the Supporting Information. Because
the voltammograms were either irreversible or at best quasi-
reversible, EHOMO was estimated from the onset potential in the
voltammogram, measured vs Ag/Ag+, and calibrated against the
ferrocene/ferrocenium couple (Fc/Fc+ measured as 0.09 V vs
Ag/Ag+). The Fc/Fc+ energy level used in HOMO calculations
was assumed to be −4.80 eV.46 Thus, EHOMO = −(ϕox + 4.71)
eV, where ϕox is the oxidation onset potential of the sample vs
Ag/Ag+. Polymer films were drop cast onto a platinum
electrode from solutions with uniform concentrations (1 mg/
mL). The data are summarized in Table 1.
Table 1. Number-Average Molecular Weight (Mn),
Polydispersity Index (PDI), and Optical and
a
Electrochemical Properties of Polymers
Mn
λmax, λonset (nm) λmax, λonset
Egopt
E
HOMdO
b
c
polymer (kDa) PDI
solution
(nm) film
(eV)
(eV)
P1
P2
P4
P5
P6
37.0
27.2
40.3
42.5
44.1
2.5
2.4
2.5
2.4
1.8
632, 665
613, 640
651, 786
630, 745
630, 752
629, 680
609, 642
662, 870
662, 842
642, 797
1.83
1.93
1.43
1.47
1.56
−5.39
−5.67
−5.37
−5.49
−5.39
e)
a
P3 was not sufficiently soluble for SEC or solution absorption
b
c
measurements. Measured in chloroform solution. Calculated from
d
film absorption onset. EHOMO estimated from onset potential
measured at 100 mV/s vs Ag/Ag+ and calibrated against Fc/Fc+
(measured as 0.09 V vs Ag/Ag+); Fc/Fc+ energy level used in HOMO
calculations was −4.80 eV.46 Voltammogram of P2 recorded at 500
e)
mV/s.
If the polymerization was allowed to proceed for 90 h at 110
°C, the polymer molecular weights obtained were higher, but
the polydispersities were quite broad (PDI > 4.0). Recent
studies suggest that optimal solar cell device efficiencies require
polymers with optimized molecular weights. Those with low
molecular weights (<10 kDa) tend to have poor transport
properties.47,48 The effects of higher molecular weights have
been less studied, but it has been demonstrated that certain
high molecular weight samples of poly-3-hexylthiophene
(P3HT) form strong aggregates in solution, ultimately leading
to poor polymer/fullerene morphologies and lower device
efficiencies at molecular weights above 150 kDa.49 Additional
studies have found that the best performance in P3HT
photovoltaic devices is obtained by using an optimum ratio
of high and low molecular weight components.50 Figure 3
illustrates how a preparatory GPC can effectively be used to
A couple of important observations can be made about these
voltammograms. First, the value of −5.39 eV estimated for P1
is close to values reported by the groups of Frechet (−5.40
́
eV)30 and Jen (−5.43 eV)29 but not quite as deep as the value
reported by the group of Leclerc (−5.56 eV).28 All three
literature values were recorded as onset potentials. Second,
substituting alkyl chain derivatives on the BDT unit for alkoxy
chains considerably lowers the measured HOMO levels,
consistent with literature observations for analogous copoly-
mers of TT and BDT.9 Those polymers with 3-ethylheptyl
substituents on BDT (P2 and P5) were estimated to have
HOMO values approximately 280 mV and 120 mV deeper than
their respective analogues with alkoxy subsituents (P1 and P4).
Additionally, the HOMO value estimated for the polymer with
the 4-butylphenethyl substituent (P6) was slightly lower than
Figure 3. GPC traces for fractions of polymer P4, synthesized at 110
°C over 90 h, after being run through a preparatory column.
1351
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
Figure 4. (a) Normalized absorbance spectra of polymers in chloroform solution. (b) Normalized absorbance spectra of polymers P2 and P5 as films
and in solution.
its alkoxy analogue (albeit, not quite as dramatically, only 20
mV). It should be noted that the onset potentials recorded for
P1 and P4−P6 were generally scan rate independent (between
100 and 500 mV/s). However, the onset potential for P2 was
200 mV more negative at 100 mV/s than it was when recorded
at 500 mV/s. Thus, the HOMO value estimated for P2 should
be deemed less reliable than values for the remaining polymers.
Figure 5 illustrates the experimental energy levels for the 5
Table 2. Device Characteristics of Photovoltaic Solar Cells
Based on Polymers P1, P2, and P4 with PC61BM
D:A
ratio
VOC
JSC (mA/
FF
PCE
(%)
Rs
polymer solvent
(mV)
cm2)
(%)
(Ω)
a
P1
oDCB
1:2
1:2
1:2
1:1
901
888
1005
792
5.6
8.0
2.0
5.0
50
67
39
52
2.5
4.8
0.8
2.1
28
b
CB
10
a
P2
P4
oDCB
b
206
15
CB
a
b
o-Dichlorobenzene. Chlorobenzene with 4% v/v diiodooctane.
Figure 5. Energy levels for polymers determined from combined
optical (film) and electrochemical (film) measurements.
polymers determined from combined optical and electro-
chemical film measurements.
Device Data. BHJ photovoltaic devices of donor polymer
P1 with acceptor PC61BM were first investigated to compare
with literature values. When devices of P1 were processed
under optimal solution conditions for that polymer, the devices
possessed a high VOC (888 mV), high FF (67%), and a
moderate JSC (8.0 mA/cm2), resulting in an overall PCE of 4.8%
(Table 2). This promising PCE is generally in good agreement
with literature results reported by Jen (4.1% with PC71BM),29
Leclerc (5.2% with PC61BM),51 and average values reported by
Fréchet (6.3% with PC61BM).30
Donor polymers P1 and P2 were then investigated to
determine how the substituents on the donor polymer would
affect device performance. The current density versus voltage
(J-V) plots of ITO/PEDOT:PSS/BHJ/Cal/Al devices are
shown in Figure 6a. The active layer blends for all polymers
investigated were spin coated from either chlorobenzene (CB)
or o-dichlorobenzene (DCB) solutions depending on solubility
Figure 6. J-V plots of ITO/PEDOT:PSS/BHJ/Ca/Al devices
comparing (a) P1 and P2 devices processed under identical conditions
from DCB solutions and (b) comparing P1 and P4 devices processed
under optimal CB solutions. (c) EQE spectra of P1 and P4 devices.
1352
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
and for proper comparison studies. The ratio of the polymers to
PC61BM in the active layer was also optimized for all polymers
for the given solutions. P2 gave optimal results when cast from
o-DCB. Comparison of P1 and P2 devices under identical
processing conditions show the VOC of P2 approximately 100
mV larger than that of P1 to give a device with a VOC greater
than one volt (Table 2). This observation is consistent with CV
measurements and literature predictions;9 both suggest that
substituting alkyl chain derivatives on the BDT unit for alkoxy
chains lowers the HOMO levels. However, the improvement in
VOC in devices of P2 came at the expense of a decreased JSC, fill
factor (FF), and PCE relative to P1 devices. Possible reasons for
the overall lower performance in P2 devices are the slightly
wider band gap of P2 leading to reduced absorption overlap
with the solar spectrum; differences in phase separation
between donor and acceptor in the BHJ due to slight
differences in solubility; and poorer hole extraction in P2
devices due to a larger offset between the HOMO of P2 and
the work function of PEDOT:PSS (∼5.0 eV52) contributing to
the poorer FF (39% versus 50%) and series resistance, Rs (206
Ω versus 28 Ω), in P2 versus P1 devices. It is possible that the
LUMO of P2 is sufficiently deep relative to that of PC61BM
that exciton dissociation becomes poor and the JSC suffers
accordingly; however, devices of P4 are estimated to have even
deeper LUMOs than P2, and yet devices of the former possess
higher values of JSC.
A comparison of P1 and P4 devices processed from their
respective optimal conditions shows that devices of P4 have a
good VOC of 792 mV but a poor JSC of 5.0 mA/cm2 and FF of
52%, ultimately leading to a PCE of only 2.1%, less than half
that of the P1 devices. The poorer JSC is despite the greater
absorption overlap of P4 with the solar spectrum versus P1 as
observed in both absorption (Figure 4) and external quantum
efficiency (EQE) measurements (Figure 6c). The relative
decrease in JSC is also not believed to be due to poorer hole
collection at the PEDOT:PSS electrode as is suspected for
devices of P2, as the HOMO values of P1 and P4 are similar
and devices exhibit comparable values of Rs (Table 2). It is
possible, if not likely, that the LUMO of P4 is sufficiently deep
relative to that of PC61BM (the latter having been estimated as
4.2 eV2) that exciton dissociation suffers. The decreased
performance in P4 devices might also be due to a packing and/
or morphology issue, which will be discussed at greater length
with computational, X-ray diffraction, and photoconductance
data below.
Figure 7. (Top and middle) Illustration of the planarity of P1 and
(bottom) twist in P4 for n = 4 oligomers.
twist throughout the backbone of TID containing polymers.
These results may help explain why little difference is observed
between the solution and film absorption spectra of P1 and P2,
while dramatic red shifts are observed for the TID containing
polymers P4-P6; in the latter case, there is much more
potential for planarity to be induced in the solid state through
favorable π−π stacking. The results are also consistent with the
X-ray diffraction and photoconductance data (vide infra).
X-ray Diffraction (XRD). X-ray diffraction measurements
were performed on pristine films of P1 and P4, as well as
blends of P1:PC61BM and P4:PC61BM with the same blending
ratios as in the optimized devices, to investigate any potential
ordering of the polymer domains in the solid state. The
diffraction patterns are shown in Figure 8. Consistent with
literature results for this polymer, the XRD pattern for P1 and
its fullerene blend exhibit a peak at 2θ = 4.20° (21.0 Å), which
has been suggested to correspond with lamellar spacing and can
often be correlated with the length of the polymer side
chains.30,51 The peak position does not shift when PC61BM is
Theoretical. HOMO levels, LUMO levels, and absorption
spectra for oligomers related to polymers P1, P2, P4, and P5
were calculated for comparison to the measured polymer
properties. The bulk of these results and discussion are
presented in the Supporting Information. Here, we briefly
discuss the geometric structures of the different polymers as
they are relevant to observations in the absorption spectra and
ordering in the polymer films.
Figure 7 illustrates the geometric structures for oligomers of
P1 and P4 as calculated by density functional theory (DFT)
and optimized in vacuum as described in the Experimental
Section. As can be seen in the top and middle of Figure 7, the
oligomer of P1 is quite planar, with a dihedral angle between
comonomers of TPD and BDT calculated as just 2−3° in
vacuum. However, in the case of P4, the hydrogen atom on the
isoindole unit of TID creates sufficient steric hindrance with the
neighboring BDT unit that a dihedral angle of 30−35° is
induced between these comonomers, introducing significant
Figure 8. X-ray diffraction patterns of pristine films of P1 and P4, as
well as P1:PC61BM and P4:PC61BM blends.
1353
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
blended in the film, suggesting that the ordered polymer
domains are not affected by the presence of the fullerene.
Interestingly, the intensity of the peak for the blend of P1 is
higher than that of the pristine P1 despite similar amounts of
polymer in the two samples. Further experiments are in
progress studying the nature of this phenomenon.
slower decay than that of P4:PC61BM. Figure 9b shows the
product of the yield for free carrier generation with the sum of
the mobilities of free carriers (ϕΣμ, see eq 1 in the
Experimental Section) as a function of absorbed photon flux
for the two blends as well as the pure polymers. The decline of
ϕΣμ at high light intensities has been documented previously53
and is attributed to quenching of excitons by free carriers, a
higher order loss mechanism that limits ϕ. In the following, the
low-intensity range of the TRMC results is discussed, where
such higher order loss mechanisms are negligible.
The low TRMC signal from the pure polymers prevented us
from obtaining photoconductance data at low excitation
intensities, as shown in Figure 9b. However, it is clear that
ϕΣμ increases in the blends compared to the pure polymers:
the presence of the polymer-fullerene interface efficiently
dissociates excitons into free carriers so that a) ϕ increases and
b) Σμ potentially also increases due to the contribution from
free electrons in PC61BM domains in the blend.43,54
At low excitation intensities the ϕΣμ for P1:PC61BM is
higher by a factor of 7 than that of P4:PC61BM. We attribute
this, in part, to a lower magnitude of Σμ in P4:PC61BM due to
lack of ordering; computational simulations (vide supra) show
that P1 acquires a planar configuration, while steric interactions
in P4 cause it to twist and become nonplanar. As seen in the
XRD, the planarity of P1 results in at least partially ordered
domains. In contrast, the nonplanarity of P4 seemingly
prevents the formation of ordered domains. The photo-
conductance decay for the two blends is also consistent with
an ordered structure, i.e. a slower decay, for P1:PC61BM and
faster for the disordered P4:PC61BM, as has been observed
previously.55 This implies higher recombination losses in the
device with a P4:PC61BM active layer because of slower
transport of photogenerated carriers in this case.
No ordering is observed in the film of pristine P4, consistent
with the twisted backbone of P4 observed in the computational
modeling that should encumber the formation of ordered
domains. The P4:PC61BM film has an extremely weak
diffraction feature around 2θ = 3°, indicating very limited
ordering in this case. The lack of ordering in the P4: PC61BM
film may, at least in part, be responsible for the lower JSC (5.0
mA/cm2) as compared to the planar P1 (JSC = 8.0 mA/cm2) in
solar cell devices, as discussed further in the next section.
Time-Resolved Microwave Conductivity (TRMC). Fig-
ure 9a shows the photoconductance transients, normalized by
absorbed photon flux and constants (see eq 1 of the
Experimental Section) for films of P1:PC61BM and
P4:PC61BM with weight ratios of 1:2 and 1:1, the same weight
ratios used in the respective optimized devices. The photo-
conductance of P1:PC61BM has both a higher magnitude and
Energetic considerations may also cause a decrease of the
yield for free carrier generation in P4:PC61BM. The LUMO of
P4 is deeper by ca. 0.38 eV than that on P1 (Figure 5), and the
corresponding decrease in the free energy available for free
carrier generation by exciton dissociation at the polymer-
fullerene interface may limit ϕ. Further experiments are in
progress to determine the extent to which P4:PC61BM devices
are limited by poor free carrier generation.
CONCLUSIONS
■
This work primarily sought to contribute to the ongoing design
effort aiming to fine-tune band gaps and HOMO levels of OPV
polymers. By considering a polymer’s resonance forms when
designing its constituent monomers, we successfully synthe-
sized and characterized several new copolymers of TID and
derivatives of BDT with appreciably lower band gaps than
analogous copolymers of TPD (lower by ∼0.4 eV). The
HOMO levels of the copolymers were then tuned with
chemical modifications to the BDT monomer, resulting in a
photovoltaic device with an open circuit voltage greater than 1
V for P2. Despite the optimized band gap, the TID containing
polymer displayed lower peak photoconductance and a faster
decay in the first 400 ns, as determined by time-resolved
microwave conductivity, and decreased device efficiency (2.1%
vs 4.8%) as compared with its TPD containing analogue. It is
likely that the LUMOs of these new materials are simply too
deep relative to that of the fullerene acceptor to efficiently
dissociate excitons in photovoltaic devices. The results may be
also partially attributed to morphological issues, as computa-
tional modeling suggests TID copolymers have a twisted
Figure 9. a) Time-Resolved Microwave Conductivity transients of thin
films of P1:PC61BM and P4:PC61BM deposited on quartz under the
same conditions as the active layers of the optimized devices presented
in Table 2 and Figure 6. The absorbed photon flux was ca. 1011
photons/cm2/pulse. Film P1:PC61BM was excited at 620 nm and
P4:PC61BM at 660 nm. b) The product of the yield for free carrier
generation and the sum of mobilities of electrons and holes obtained
from the peak photoconductance, as a function of absorbed photon
flux.
1354
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
(18) Roncali, J. Macromol. Rapid Commun. 2007, 28 (17), 1761−
1775.
(19) Ajayaghosh, A. Chem. Soc. Rev. 2003, 32 (4), 181−191.
(20) Wudl, F.; Kobayashi, M.; Heeger, A. J. J. Org. Chem. 1984, 49
(18), 3382−3384.
(21) Ikenoue, Y.; Wudl, F.; Heeger, A. J. Synth. Met. 1991, 40 (1), 1−
12.
backbone, and X-ray diffraction data indicate films of this
polymer do not form ordered domains, whereas TPD
copolymers are considerably more planar and form partially
ordered domains. Given these results, we did not continue to
study devices of P5 and P6, which also possess deep LUMOs
and were invariably subject to similar morphological issues;
rather, we are focusing our efforts on making more planar
derivatives of TID copolymers.
(22) Meng, H.; Chen, Y.; Wudl, F. Macromolecules 2001, 34 (6),
1810−1816.
(23) Pomerantz, M.; Chalonergill, B.; Harding, L. O.; Tseng, J. J.;
Pomerantz, W. J. Synth. Met. 1993, 55 (2−3), 960−965.
(24) Sotzing, G. A.; Lee, K. H. Macromolecules 2002, 35 (19), 7281−
7286.
(25) Hou, J. H.; Park, M. H.; Zhang, S. Q.; Yao, Y.; Chen, L. M.; Li, J.
H.; Yang, Y. Macromolecules 2008, 41 (16), 6012−6018.
(26) Hou, J. H.; Chen, H. Y.; Zhang, S. Q.; Chen, R. I.; Yang, Y.; Wu,
Y.; Li, G. J. Am. Chem. Soc. 2009, 131 (43), 15586−15587.
(27) Liang, Y. Y.; Xu, Z.; Xia, J. B.; Tsai, S. T.; Wu, Y.; Li, G.; Ray, C.;
Yu, L. P. Adv. Mater. 2010, 22 (20), E135.
(28) Zou, Y. P.; Najari, A.; Berrouard, P.; Beaupre, S.; Aich, B. R.;
Tao, Y.; Leclerc, M. J. Am. Chem. Soc. 2010, 132 (15), 5330.
(29) Zhang, Y.; Hau, S. K.; Yip, H. L.; Sun, Y.; Acton, O.; Jen, A. K.
Y. Chem. Mater. 2010, 22 (9), 2696−2698.
(30) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.;
Beaujuge, P. M.; Frechet, J. M. J. J. Am. Chem. Soc. 2010, 132 (22),
7595.
ASSOCIATED CONTENT
■
S
* Supporting Information
Cyclic voltammograms for all polymers, absorption spectrum of
polymer films, discussion of theoretical methods. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: Wade.Braunecker@nrel.gov.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the U.S. Department of Energy
under Contract No. DE-AC36-08-GO28308 with the National
Renewable Energy Laboratory through the DOE SETP
program.
(31) Sonmez, G.; Meng, H.; Wudl, F. Chem. Mater. 2003, 15 (26),
4923−4929.
(32) Zhang, Q. T.; Tour, J. M. J. Am. Chem. Soc. 1998, 120 (22),
5355−5362.
(33) Zhang, Q. T.; Tour, J. M. J. Am. Chem. Soc. 1997, 119 (41),
9624−9631.
REFERENCES
■
(34) Reddy, P. Y.; Kondo, S.; Toru, T.; Ueno, Y. J. Org. Chem. 1997,
62 (8), 2652−2654.
(1) Gunes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107
(4), 1324−1338.
(35) Hsu, D. T.; Lin, C. H. J. Org. Chem. 2009, 74 (23), 9180−9187.
(36) Meng, H.; Tucker, D.; Chaffins, S.; Chen, Y. S.; Helgeson, R.;
Dunn, B.; Wudl, F. Adv. Mater. 2003, 15 (2), 146−149.
(37) Beimling, P.; Kossmehl, G. Chem. Ber. 1986, 119 (10), 3198−
3203.
(2) Thompson, B. C.; Frechet, J. M. J. Angew. Chem., Int. Ed. 2008, 47
(1), 58−77.
(3) Cheng, Y. J.; Yang, S. H.; Hsu, C. S. Chem. Rev. 2009, 109 (11),
5868−5923.
(4) Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science
1995, 270 (5243), 1789−1791.
(38) Gilbert, J. C.; Weerasooriya, U. J. Org. Chem. 1982, 47 (10),
1837−1845.
(5) Peet, J.; Heeger, A. J.; Bazan, G. C. Acc. Chem. Res. 2009, 42 (11),
1700−1708.
(39) Nielsen, K. T.; Bechgaard, K.; Krebs, F. C. Macromolecules 2005,
38 (3), 658−659.
(6) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009, 21
(13), 1323−1338.
(40) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.,
Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.;
Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;
Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.;
Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;
Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador,
P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
Revision B.01; Gaussian, Inc.: Wallingford, CT, 2010.
(41) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y.
Adv. Funct. Mater. 2006, 16, 2016−2023.
(7) Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.;
Neagu-Plesu, R.; Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M. J.
Am. Chem. Soc. 2008, 130 (2), 732−742.
(8) Huo, L. J.; Hou, J. H.; Chen, H. Y.; Zhang, S. Q.; Jiang, Y.; Chen,
T. L.; Yang, Y. Macromolecules 2009, 42 (17), 6564−6571.
(9) Liang, Y. Y.; Feng, D. Q.; Wu, Y.; Tsai, S. T.; Li, G.; Ray, C.; Yu,
L. P. J. Am. Chem. Soc. 2009, 131 (22), 7792−7799.
(10) Huo, L. J.; Hou, J. H.; Zhang, S. Q.; Chen, H. Y.; Yang, Y.
Angew. Chem., Int. Ed. 2010, 49 (8), 1500−1503.
(11) Brabec, C. J.; Cravino, A.; Meissner, D.; Sariciftci, N. S.;
Fromherz, T.; Rispens, M. T.; Sanchez, L.; Hummelen, J. C. Adv.
Funct. Mater. 2001, 11, 374−380.
(12) Scharber, M. C.; Wuhlbacher, D.; Koppe, M.; Denk, P.;
Waldauf, C.; Heeger, A. J.; Brabec, C. L. Adv. Mater. 2006, 18 (6),
789−794.
(13) Bundgaard, E.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2007, 91
(11), 954−985.
(14) Osaka, I.; Zhang, R.; Sauve, G.; Smilgies, D. M.; Kowalewski, T.;
McCullough, R. D. J. Am. Chem. Soc. 2009, 131 (7), 2521−2529.
(15) Cates, N. C.; Gysel, R.; Beiley, Z.; Miller, C. E.; Toney, M. F.;
Heeney, M.; McCulloch, I.; McGehee, M. D. Nano Lett. 2009, 9 (12),
4153−4157.
(16) http://optics.org/indepth/2/12/2 (accessed March 2012).
(17) van Mullekom, H. A. M.; Vekemans, J. A. J. M.; Havinga, E. E.;
Meijer, E. W. Mater. Sci. Eng. 2001, 32 (1), 1−40.
(42) Kroeze, J. E.; Savenije, T. J.; Vermeulen, M. J. W.; Warman, J.
M. J. Phys. Chem. B 2003, 107, 7696−7705.
(43) Ferguson, A. J.; Kopidakis, N.; Shaheen, S. E.; Rumbles, G. J.
Phys. Chem. C 2011, 115 (46), 23134−23148.
(44) Dayal, S.; Kopidakis, N.; Rumbles, G. Faraday Discuss. 2012, in
press.
(45) Park, J. K.; Jo., J.; Seo., J. H.; Moon, J. S.; Park, Y. D.; Lee, K.;
Heeger, A. J.; Bazan, G. C. Adv. Mater. 2011, 23, 2430−2435.
1355
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356

Chemistry of Materials
Article
(46) Pommerehne, J.; Vestweber, H.; Guss, W.; Mahrt, R. F.; Bassler,
H.; Porsch, M.; Daub, J. Adv. Mater. 1995, 7 (6), 551−554.
(47) Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C.
J. Chem. Mater. 2005, 17 (8), 2175−2180.
(48) Tong, M. H.; Cho, S.; Rogers, J. T.; Schmidt, K.; Hsu, B. B. Y.;
Moses, D.; Coffin, R. C.; Kramer, E. J.; Bazan, G. C.; Heeger, A. J. Adv.
Funct. Mater. 2010, 20 (22), 3959−3965.
(49) Koppe, M.; Brabec, C. J.; Heiml, S.; Schausberger, A.; Duffy, W.;
Heeney, M.; McCulloch, I. Macromolecules 2009, 42 (13), 4661−4666.
(50) Ma, W.; Kim, J. Y.; Lee, K.; Heeger, A. J. Macromol. Rapid
Commun. 2007, 28 (17), 1776−1780.
(51) Najari, A.; Beaupre, S.; Berrouard, P.; Zou, Y. P.; Pouliot, J. R.;
Lepage-Perusse, C.; Leclerc, M. Adv. Funct. Mater. 2011, 21 (4), 718−
728.
(52) Koch, N.; Kahn, A.; Ghijsen, J.; Pireaux, J.-J.; Schwartz, J.;
Johnson, R. L.; Elschner, A. Appl. Phys. Lett. 2003, 82, 70−72.
(53) Ferguson, A. J.; Kopidakis, N.; Shaheen, S. E.; Rumbles, G. J.
Phys. Chem. C 2008, 112, 9865−9871.
(54) Savenije, T. J.; Kroeze, J. E.; Wienk, M. M.; Kroon, J. M.;
Warman, J. M. Phys. Rev. B 2004, 69, 155205.
(55) Rance, W. R.; Rupert, B. L.; Mitchell, W. J.; Kose, M. E.; Ginley,
̈
D. S.; Shaheen, S. E.; Rumbles, G.; Kopidakis, N. J. Phys. Chem. C
2010, 114, 22269−22276.
1356
dx.doi.org/10.1021/cm2038427 | Chem. Mater. 2012, 24, 1346−1356