Improved synthesis of thienothiazole and its utility in developing polymers for photovoltaics

Article abstract of DOI:10.1021/ma201835h

In response to the structural and electronic limitations of the popular benzo[1,2-b:4,5-b′]dithiophene-thieno[3,4-b]thiophene (PBnDT-TT) polymer series, this study explores the design and synthesis of a thienothiazole (TTz) moiety. The synthesis of TTz was streamlined down to four high-yielding steps, resulting in the new polymer PBnDT-TTz for organic solar cells. By incorporating TTz, a nitrogen is directly introduced into the polymer backbone which tunes the HOMO level and eliminates the reliance on external substituents. Compared to its TT analogue, PBnDT-TTz exhibits the same HOMO level of -5.06 eV and the same V_oc of 0.69 eV, yet a higher power conversion efficiency of 2.5%. These promising results demonstrate the benefits of backbone modification and the great potential of TTz in the design of new polymers for organic photovoltaics.

Full text of DOI:10.1021/ma201835h

ARTICLE
pubs.acs.org/Macromolecules
Improved Synthesis of Thienothiazole and Its Utility in Developing
Polymers for Photovoltaics
Rycel Uy,† Liqiang Yang,‡ Huaxing Zhou,† Samuel C. Price,† and Wei You*,†,‡
†Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States
‡Curriculum in Applied Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3287,
United States
S Supporting Information
b
ABSTRACT: In response to the structural and electronic
limitations of the popular benzo[1,2-b:4,5-b0]dithiopheneꢀ
thieno[3,4-b]thiophene (PBnDTꢀTT) polymer series, this
study explores the design and synthesis of a thienothiazole
(TTz) moiety. The synthesis of TTz was streamlined down to
four high-yielding steps, resulting in the new polymer PBnDTꢀ
TTz for organic solar cells. By incorporating TTz, a nitrogen is
directly introduced into the polymer backbone which tunes the HOMO level and eliminates the reliance on external substituents.
Compared to its TT analogue, PBnDTꢀTTz exhibits the same HOMO level of ꢀ5.06 eV and the same Voc of 0.69 eV, yet a higher
power conversion efficiency of 2.5%. These promising results demonstrate the benefits of backbone modification and the great
potential of TTz in the design of new polymers for organic photovoltaics.
’ INTRODUCTION
example, Hou et al. replaced the ester with a ketone but observed
almost no change in the HOMO level.15 Similarly, when fluorine
was placed on TT, this lowered the HOMO level by only
0.11 eV.14 When Huang et al. used the even more electronegative
sulfonyl on TT instead, the resulting HOMO level was still very
similar.17 Although adding and changing substituents has shown
some success, directly engineering the polymer backbone is a
more effective method to tune energy levels.22,27 Given that
further structural modifications to TT have been nearly ex-
hausted, a new unit needs to be designed that still incorporates
the advantages of TT while offering similar or lower HOMO
levels in its polymers.
Although substantial improvements have been made in the
development of donor polymers for organic solar cells1ꢀ5 reach-
ing over 5% power conversion efficiency,6ꢀ12 recent years have
seen only incremental advances. This highlights the need to
probe the inherent issues limiting current donor materials and
focus on discovering design fundamentals necessary to develop
better-performing polymers. Currently, one of the top perfor-
mers is a polymer series consisting of alternating benzo[1,2-
b:4,5-b0]dithiophene (BnDT) and thieno[3,4-b]thiophene (TT)
moieties that have reached efficiencies over 7%.13ꢀ18
The success of the PBnDTꢀTT series can be attributed to this
material meeting four of the five generally accepted criteria
necessary for donor polymers to perform well in bulk hetero-
junction (BHJ) solar cells.16,19ꢀ22 Because the TT moiety can
stabilize its quinoidal form, this leads to more double-bond
character in the polymer backbone which leads to (1) a low
band gap of ∼1.5 eV and (2) high hole mobility. Also, since every
BnDTꢀTT unit contains three solubilizing side groups, the alkyl
chains need not be excessively long. Moderate chain lengths not
only allow PBnDTꢀTT to achieve (3) good solubility in spin-
casting solvents and (4) high molecular weight23,24 but also help
to enhance short circuit current (Jsc).25 However, even with
an electron-withdrawing ester as the stabilizing group, the
electron-rich nature of TT raises the highest occupied mole-
cular orbital (HOMO) level too high at about ꢀ5.0 eV,14,26
failing to achieve (5) a low-lying HOMO level around ꢀ5.4 eV
and leading to lower open circuit voltage (Voc) and therefore
lower efficiency.3,21
One way to increase the electron-deficiency of TT and eli-
minate the need for external ester or fluorine substituents is to
introduce an electron-deficient unit such as thiazole. This will
change the thienothiophene (TT) to thienothiazole (TTz),
which is also capable of stabilizing its quinoid form.28 To gauge
the effect on the HOMO level, density functional theory (DFT)
calculations were collected for alkylated TTz and TTs with an
alkyl or an ester alkyl chain (Figure 1). All three were paired
with BnDT29,30 since this unit has demonstrated numerous
favorable results in BHJ solar cells.10,16 As shown in Table 1,
calculations predict that adding the ester to the TT versus
changing the carbon to nitrogen (TTz) lower the HOMO level
by a similar amount, 0.15 and 0.14 eV, respectively. Thus, in
this initial study we want to investigate how the TTz-based
polymer’s solar cell performance will compare with that of the
ester TT-based polymer.
Received: August 10, 2011
In an effort to lower the HOMO level of the polymer, TT has
been substituted with various electron-withdrawing groups. For
Revised:
September 27, 2011
Published: November 03, 2011
r
2011 American Chemical Society
9146
dx.doi.org/10.1021/ma201835h Macromolecules 2011, 44, 9146–9154
|

Macromolecules
ARTICLE
Figure 1. Oligomeric units of (A) alkyl TT, (B) ester TT, and (C) alkyl TTz used to perform DFT calculations.
Table 1. DFT Calculations for BnDT Copolymerized with
Alkylated TT, Ester TT, and Alkylated TTz
’ RESULTS AND DISCUSSION
Synthesis of Thienothiazole Monomer and Polymer. The
originally published procedure28 was reproduced on a test run
using an undecyl alkyl chain with the following yields shown in
Scheme 1. As can be seen, this synthetic route suffers from several
drawbacks. The most significant one is that the chain length must
be decided in the first step, making later structural modifications
to TTz inefficient and tedious. Furthermore, seven steps are
required to get to TTz, many of which are low-yielding, especially
in the later steps forming the thiophene ring.
HOMO [eV]
LUMO [eV]
Eg [eV]
alkyl TT
ester TT
alkyl TTz
ꢀ5.06
ꢀ5.21
ꢀ5.20
ꢀ1.75
ꢀ2.17
ꢀ1.86
3.31
3.04
3.34
Rather than starting with thiazole and attempting to form the
second thiophene ring, we decided that starting with the thio-
phene ring would be more advantageous, since a variety of func-
tionalized thiophenes are commercially available. Additionally,
carbonꢀheteroatom bond disconnections allow for a much faster
synthesis and a broader scope of reactions to be considered.33
Thus, we decided to construct the thiazole from an amide at
the 3-position of the thiophene and then close the ring via
Cu-catalyzed cyclization between the carbonꢀsulfur bond.34
Functionalizing the thiophene with an amide at this position
provides us with a number of benefits (Figure 3). It allows us to
anchor on our desired alkyl chain, which can later be converted to
a thioamide easily, and there are a multitude of amide precursors
we can start from such as a nitro group, azide, amine, carboxylic
acid, etc.
We chose to begin with a carboxylic acid and mapped out an
alternate synthetic route (Scheme 2). However, 3-bromothio-
phene-4-carboxylic acid (1) itself could not be generated pure
and in high yield. Solvent choice played an important role in this
halogenꢀlithium exchange reaction. Tetrahydrofuran could not
be used since 3,4-dibromothiophene can undergo scrambling in
this solvent.35 When hexanes was used, the desired carboxylic
acid was isolated but in low yield since 3,4-dibromothiophene is
only sparingly soluble in hexanes at low temperatures. There was
no solubility issue when diethyl ether was used as the solvent, and
furthermore, deuterium-labeling experiments proved that scram-
bling of the starting material did not occur in this solvent. How-
ever, this reaction in diethyl ether led to the dicarboxylic acid as the
major product (Figure 4), most likely because 3-bromothiophene-
4-carboxylic acid is more reactive toward halogenꢀlithium
exchange than 3,4-dibromothiophene.
Figure 2. Two different approaches to tune the energy levels of thie-
nothiophene, where dashed arrows indicate further structural variations.
In addition to a similar HOMO level, TTz possesses several
potential improvements over ester TT. The unfunctionalized α-
position provides a synthetic handle for further modifications
that would allow for further lowering of the HOMO level
(Figure 2). To date, there had only been one previous report
of thienothiazole in the literature,28 which involved seven mostly
low-yielding steps. This procedure was recently reproduced by
Leclerc and co-workers during the preparation of this paper, and
the reported yields were similarly low.31 If the number of steps
could be drastically shortened, TTz would also be synthetically
favorable over TT which can take four to seven steps, each
rendition reported as being difficult and costly.14,15,17,18,32 These
synthetic challenges with TT will hinder this material’s viability
of being reproduced on a commercial scale in the future.
Herein we report a new, facile synthetic route to TTz, involving
only four high-yielding steps. To make TTz, the key bond disconnec-
tion is the CꢀS bond in the top thiazole ring rather than the CꢀSꢀC
bond formation toward the bottom thiophene unit, allowing us to
make the unit much faster. From this optimized route, a new polymer
PBnDTꢀTTz was made and its properties were studied alongside its
TT analogue. Compared to PBnDTꢀTT, PBnDTꢀTTz exhibits
better surface morphology, hole mobility, Jsc, and thus higher solar cell
efficiency (2.3% over 1.2%), demonstrating the great potential of this
new structural moiety (TTz) in organic photovoltaics.
A third route shown in Scheme 3 was then proposed. Rather
than synthesizing a boc-protected amine and then deprotecting
to attach the alkyl chain as was originally suggested in Scheme 2,
we decided to make 3-azido-4-bromothiophene36 and reduce it
to the free amine (2) in one pot. The corresponding acid chloride
is then added to attach the alkyl chain of choice (ethylhexyl) to
9147
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
Scheme 1. Synthesis of Thienothiazole, Procedure I, Reproduction of Literature Procedure with Observed Yields
Figure 3. Retrosynthetic analysis of thienothiazole.
Scheme 2. Synthesis of Thienothiazole, Procedure II, via
3-Bromothiophene-4-carboxylic Acid
Figure 4. Proton NMR spectra of (A) 3,4-dibromothiophene, (B)
3-bromothiophene-4-carboxylic acid, and (C) 3,4-dicarboxylic acid
thiophene.
obtain our amide (3). Particularly noteworthy in this synthesis is
the Cu-catalyzed cyclization34 of the thioamide (4), which was
adapted from a procedure reported by Evindar et al.34 This CꢀS
bond forming reaction is very clean, proceeds in quantitative
yield, and has previously been demonstrated as an efficient way to
form benzothiazoles34 and now thienothiazoles as well. Proce-
dure III was our first successful synthesis of TTz and was an
improvement over the previous two routes in that the yields were
higher and that it afforded the target compound in only five steps.
In order to further optimize the route, we decided to isolate
3-azido-4-bromothiophene instead and treat it with the appro-
priate carboxylic acid to attach the alkyl chain of choice (ethylhexyl).
This reaction is known as the StaudingerꢀVilarrasa reaction,37
which allows us to synthesize our desired amide from the azide in
one step. This synthetic route (Scheme 4) is a dramatic improve-
ment over the previous methods since it has only four steps, yields are
higher, and the chain length is introduced much later in the synthesis.
TTz is then brominated via N-bromosuccinimide (NBS) (5) and
polymerized via Stille polycondensation38,39 with the corre-
sponding benzodithiophene distannane monomer (BnDT)
using conventional heating conditions to afford the new
PBnDTꢀTTz polymer (Scheme 5).
For the sake of comparison, the analogous TT polymer was
also synthesized according to the literature14 since the photo-
voltaic (PV) properties of this specific BnDTꢀTT polymer with
ethylhexyl chains on both units had not been previously reported.
Both polymers were purified by Soxhlet extraction with metha-
nol, ethyl acetate, hexanes, and chloroform. Using the resulting
solids from the chloroform fraction, the molecular weights of the
polymers were determined by high-temperature gel permeation
chromatography (GPC) in 1,2,4-trichlorobenzene at 135 °C
(Table 2). For conventional heating, both polymers yielded
comparable and moderately high molecular weights.
Optical and Electrochemical Properties. The absorption
spectra and the cyclic voltammograms of the polymers are shown
9148
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
in Figure 5, with all the representative data summarized in
Table 3. Compared to PBnDTꢀTT, PBnDTꢀTTz exhibits a
more pronounced absorption shoulder at 628 nm, indicating strong
aggregation (πꢀπ stacking) in solution. However, PBnDTꢀTT
has a broader absorption and a smaller band gap by 0.24 eV, most
likely due to TT’s lower resonance energy. According to DFT
calculation results shown in Table 4, the difference between the
band gaps of the aromatic and quinoid ester BnDTꢀTT is much
smaller (1.33 eV) than BnDTꢀTTz (1.58 eV). This suggests that
TT tends to favor its quinoid form more so than TTz does, leading
to PBnDTꢀTT’s smaller band gap. Nevertheless, the observed
HOMO levels40 (estimated by electrochemical measurement) of
both polymers are identical (ꢀ5.06 eV) as the calculations
predicted (Table 1), which implies that the introduction of the
nitrogenwithinthepolymer backboneindeedhasa similareffecton
the energy levels as an ester substituent does.
Photovoltaic Properties. Despite having the same HOMO
level, PBnDTꢀTTz and PBnDTꢀTT display vastly different PV
properties (Table 5). The first major difference is the influence
of the spin-casting solvent used, either chlorobenzene (CB) or o-di-
chlorobenzene (DCB). For PBnDTꢀTT, solvent choice made little
impact on its PV properties. PCE, mobility, Voc, Jsc, fill factor, and
absorption coefficient values were all very similar in both CB and
DCB (Table 5, Figures 6 and 7). Although these values differ from
the polymer series reported by Liang et al.,13,14,26 it is important to
note that this specific side chain combination has not been reported,
and it is not surprising that a slight modification in chain length would
yield such a difference in PV performance.25,41,42
PBnDTꢀTTz, on the other hand, is greatly influenced by the
processing solvent. In DCB, PV properties are inferior, especially
Jsc and the absorption coefficient. As discussed previously,
PBnDTꢀTTz has a stronger πꢀπ stacking ability than PBnDTꢀ
TT. Therefore, in the high boiling solvent (DCB) which extends
the solvent annealing time, PBnDTꢀTTz chains have more time
to agglomerate, detrimentally leading to further phase separation
between polymer chains and PCBM molecules. This is also
confirmed in the atomic force microscopy (AFM) images in
which the surface morphology of PBnDTꢀTTz exhibits large
domains in DCB (Figure 8D) compared to CB (Figure 8C).
These strong agglomerations of PBnDTꢀTTz produce a non-
uniform PBnDTꢀTTz/PCBM thin film processed in DCB and,
consequently, a low absorption coefficient and Jsc. Low boiling
solvent (CB) shortens annealing time and therefore reduces
excessive polymerꢀPCBM segregation, leading to a more favor-
able surface morphology in the PBnDTꢀTTz/PCBM thin film
(Figure 8C). In this initial study, AFM was used to survey the
surface nature of the polymer/PCBM films which are rather thin
at ∼100 nm (Table 5). However, more advanced techniques
such as defocused transmission electron microscopy (TEM) or
neutron scattering will be used in the future to determine the
actual bulk morphology of PBnDTꢀTTz/PCBM films.
Scheme 3. Synthesis of Thienothiazole, Procedure III, via
3-Amino-4-bromothiophene
Scheme 4. Synthesis of Thienothiazole, Procedure IV, via
3-Azido-4-bromothiophene
When processed in CB, both polymers exhibit the same Voc
(0.69 eV), further signifying that changing to the thiazole unit has
a similar effect as substituting TT with an ester. The absence of
the external ester in PBnDTꢀTTz case appears to have led to
smaller surface domains, higher fill factor (42.4% vs 35.1%),
higher Jsc (8.64 mA/cm2 vs 4.99 mA/cm2), and higher hole
Table 2. Polymerization Results for Polymers
Mn [kg/mol]
Mw [kg/mol]
PDI
yield [%]
PBnDTꢀTTz
PBnDTꢀTT
25.7
16.6
62.6
43.3
2.43
2.61
90.3
35.7
Scheme 5. Synthesis of Polymers (A) PBnDTꢀTTz and (B) PBnDTꢀTT
9149
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
Figure 5. (A) UV-vis absorption spectra and (B) cyclic voltammograms of PBnDTꢀTTz and PBnDTꢀTT.
Table 3. Optical and Electrochemical Data of Polymers
UVꢀvis absorption
cyclic voltammetry
CHCl3 solution
film
Eox,onset (V)
Ered,onset (V)
λmax [nm]
λonset [nm]
Eg [eV]
λmax [nm]
λonset [nm]
Eg [eV]
HOMO [eV]
LUMO [eV]
PBnDTꢀTTz
PBnDTꢀTT
628
670
658
750
1.88
1.65
632
673
670
766
1.85
1.61
ꢀ5.06
ꢀ5.06
ꢀ2.82
ꢀ3.14
mobility (4.52 ꢁ 10ꢀ5 vs 2.18 ꢁ 10ꢀ4), ultimately making this
polymer more efficient.
Table 4. DFT Calculations of Quinoid Structures for BnDT
Copolymerized with Alkyl TT, Ester TT, and Alkyl TTz
Eg difference vs
’ CONCLUSIONS
HOMO [eV]
LUMO [eV]
Eg [eV]
aromatic [eV]
In summary, we have established a new synthetic route to
thienothiazole, streamlining it down from seven to four much
higher yielding steps. Furthermore, side-chain modification can
be done much later in the synthesis, making it easier to create a
library of thienothiazole-based compounds. From this new route,
a new PBnDTꢀTTz polymer was made and compared to its TT
counterpart, which is synthetically harder to produce. Both
polymers exhibit the same Voc and HOMO level, indicating that
incorporating a nitrogen into the backbone can achieve a similar
effect without sacrificing TT’s beneficial qualities and eliminates
the reliance on external substituents. Interestingly, PBnDTꢀTTz
demonstrated solvent dependence, exhibiting far better surface
morphology and solar cell performance in CB than in DCB.
Despite having a larger band gap than its TT analogue, PBnDTꢀ
TTz’s Jsc and mobility are considerably higher, leading to higher
solar cell efficiency. These favorable results support the design
motif that backbone modification can be more advantageous
than an external substituent, and it also demonstrates the use-
fulness of the TTz moiety in photovoltaics. Furthermore, be-
cause the TTz used in this initial study has only a simple alkyl
chain, other structural modifications can be made in the future to
improve solar cell performance. Such studies are currently
underway and will be reported in due course.
alkyl TT
ester TT
alkyl TTz
ꢀ4.70
ꢀ4.89
ꢀ4.78
ꢀ2.94
ꢀ3.18
ꢀ3.02
1.76
1.71
1.76
1.55
1.33
1.58
Anhydrous diethyl ether, toluene, and DMF were used as received. THF
was dried by distillation from sodium/benzophenone. All other chemi-
cals were purchased from commercial sources (Acros, Alfa Aesar, Sigma-
Aldrich, Fisher Scientific, Oakwood Chemical, Matrix Scientific) and
used without further purification.
N-(4-Bromothiophenyl)-3-ethylheptanamide (3). To a mix-
ture of PySSPy (3.16 mmol, 0.696 g) and carboxylic acid (15.8 mmol,
2.50 g) stirring in anhydrous toluene (125 mL) under argon, 2-azido-3-
bromothiophene (15.8 mmol, 3.23 g) was added. 1.0 M trimethylpho-
sphine solution in toluene (48.7 mmol, 48.7 mL) was then slowly added
at 0 °C. Reaction mixture was allowed to warm to room temperature and
stirred overnight. Afterward, the mixture was extracted with saturated
sodium bicarbonate and ethyl acetate. The organic layers were combined
and dried over MgSO4. The compound was then purified by column
chromatography by eluting with 5% ethyl acetate/hexanes to afford a
colorless oil. Yield: 65%. 1H NMR (CDCl3, 300 MHz, δ): 7.91 (d, J =
3.6 Hz, 1H), 7.23 (d, J = 3.6 Hz, 1H), 2.33 (d, J = 6.9 Hz, 2H), 1.92
(m, 1H), 1.34 (m, 8H), 0.90 (m, 6H). 13C NMR (CDCl3, 400 MHz, δ):
170.01, 132.58, 121.22, 110.37, 104.20, 41.88, 36.91, 33.00, 28.84, 26.24,
22.92, 14.05, 10.84.
N-(4-Bromothiophenyl)-3-ethylheptanethioamide (4).
Lawesson’s reagent (15.3 mmol, 6.19 g) was added to a solution of
compound (3) (10.2 mmol, 2.96 g) stirring in anhydrous THF (125 mL).
The reaction mixture was refluxed overnight. Afterward, the reaction
’ EXPERIMENTAL SECTION
Reagents. 2-Azido-3-bromothiophene36 was synthesized according
to the literature. All solvents are ACS grade unless otherwise noted.
9150
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
Table 5. Characteristic Photovoltaic Data for PBnDTꢀTTz and PBnDTꢀTT
polymer
solvent
thickness [nm]
mobility [cm2/(V s)]
Jsc [mA/cm2]
Voc [V]
FF [%]
PCE [%]
PBnDTꢀTT CB
PBnDTꢀTT DCB
PBnDTꢀTTz CB
PBnDTꢀTTz DCB
105
94
4.52 ꢁ 10ꢀ5
2.22 ꢁ 10ꢀ4
2.18 ꢁ 10ꢀ4
4.84 ꢁ 10ꢀ4
4.99
4.96
8.64
2.73
0.69
0.69
0.69
0.63
35.1
36.6
42.4
41.3
1.2
1.3
97
2.5
112
0.71
Figure 6. Characteristic JꢀV curves of PBnDTꢀTTz and PBnDTꢀTT BHJ solar cells in (A) chlorobenzene and (B) dichlorobenzene.
Figure 7. Incident photon to current efficiency and solid film absorption of each blend of polymer:PC61BM in (A) chlorobenzene and (B)
dichlorobenzene.
mixture was partitioned between ethyl acetate and 10% NaOH solution.
The organic layer was collected and dried over MgSO4. The mixture was
then purified via column chromatography using a 5% ethyl acetate/
hexanes mixture as the eluent, resulting in a pale yellow oil. Yield: 91%.
1H NMR (CDCl3, 300 MHz, δ): 8.90 (d, J = 2.7 Hz, 1H), 7.22 (d, J =
2.7 Hz, 1H), 2.73 (d, J = 6.9 Hz, 2H), 2.03 (m, 1H), 1.33 (m, 8H), 0.88
(m, 6H). 13C NMR (CDCl3, 400 MHz, δ): 202.42, 133.67, 121.99,
114.80, 105.36, 53.92, 40.66, 32.58, 29.21, 26.06, 23.36, 14.60, 11.23.
2-(2-Ethylhexyl)thieno[3,4-d]thiazole (TTz). CuI (0.464 mmol,
88.0 mg), neocuproine (0.927 mmol, 0.193 g), and K2CO3 (13.9 mmol,
1.92 g) were quickly added to a solution of compound (4) (9.27 mmol,
3.10 g) stirring in anhydrous THF (125 mL) under argon. The reaction
mixture was then refluxed overnight and then extracted using ethyl acetate
and water. Column chromatography was performed using 5% ethyl acetate/
400 MHz, δ): 178.25, 160.02, 134.29, 130.84, 109.34, 39.86, 39.60, 32.64,
28.75, 25.86, 22.93, 14.07, 10.73.
4,6-Dibromo-2-(2-ethylhexyl)thieno[3,4-d]thiazole (5). To
a solution of thienothiazole (TTz) (9.67 mmol, 2.45 g) stirring in ACS
grade THF (125 mL) at 0 °C, N-bromosuccinimide (19.34 mmol,
3.44 g) was slowly added portion-wise. The reaction mixture was then
allowed to warm to room temperature and stirred overnight. After
confirming by TLC that the reactant was no longer present, the mixture
was then partitioned between ethyl acetate and saturated sodium
bicarbonate. The combined organic layers were then dried over MgSO4.
The compound was then isolated by column chromatography using 5%
1
ethyl acetate/hexanes, resulting in a yellow oil. Yield: 62%. H NMR
(CDCl3, 300 MHz, δ): 2.91 (d, J = 7.2 Hz, 2H), 1.86 (m, 1H), 1.44 (m,
8H), 0.85 (m, 6H). 13C NMR (CDCl3, 400 MHz, δ): 179.79, 156.75,
135.81, 128.80, 95.11, 39.93, 39.92, 32.58, 28.69, 25.86, 22.89, 14.06, 10.72.
Representative Stille Coupling Polymerization Procedure.
To a degassed solution of monomer (5) (0.516 mmol, 0.212 g) and
1
hexanes to afford a yellow oil in quantitative yield. H NMR (CDCl3,
300 MHz, δ): 7.46 (d, J = 2.7 Hz, 1H), 7.17 (d, J = 2.7 Hz, 1H), 2.92 (d, J =
7.2 Hz, 2H), 1.59 (m, 1H), 1.42 (m, 8H), 0.91 (m, 6H). 13CNMR(CDCl3,
9151
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
Figure 8. AFM images of PBnDTꢀTTz in (A) chlorobenzene and (B) dichlorobenzene.
distannylated benzodithiophene (0.516 mmol, 0.398 g) in anhydrous
DMF (2 mL) and anhydrous toluene (8 mL) under argon, tetrakis-
triphenylphosphine palladium (0.021 mmol, 24.0 mg) was quickly
added. Reaction mixture was then refluxed for 48 h. Afterward, the
reaction mixture was then precipitated into methanol. The polymer
solids were then collected and Soxhlet extracted with methanol, ethyl
acetate, hexanes, and then chloroform. The chloroform extracts were
then concentrated and precipitated into methanol and then filtered and
washed with methanol. The residual solvent was then removed under
vacuum, affording PBnDT-TTz as a gold-black solid.
Bioanalytical Systems (BAS) Epsilon potentiostat equipped with a
three-electrode configuration: a glassy carbon working electrode, a
Ag/AgNO3 (0.01 M in anhydrous acetonitrile) reference electrode,
and a Pt wire counter electrode. Measurements were taken under an
argon atmosphere at a scan rate of 100 mV/s in anhydrous acetonitrile
with tetrabutylammonium hexafluorophosphate (0.1 M) as the support-
ing electrolyte. Polymer films were drop cast onto the glassy carbon
electrode from a concentrated polymer solution in chloroform and dried
under a stream of argon prior to obtaining measurements. The potential
of the Ag/AgNO3 reference electrode was internally calibrated by using
the ferrocene/ferrocenium redox couple (Fc/Fc+). The electrochemical
onsets were determined at the position where the current starts to
deviate from the baseline. The highest occupied molecular orbital
(HOMO) and lowest unoccupied molecular orbital (LUMO) energy
levels were calculated from the onset oxidation potential (Eox) and
onset reductive potential (Ered), respectively, according to the following
equations:
Instrumentation. 1H nuclear magnetic resonance (NMR) spectra
were obtained at 300 or 400 MHz as solutions in CDCl3, 1,1,2,2-
tetrachloroethane-d2, or DMSO-d6. 13C NMR spectra were obtained at
100 MHz as solutions in CDCl3. Chemical shifts are reported in parts
per million (ppm, δ) and referenced from trimethylsilane. Coupling
constants J are reported in hertz. Spectral splitting patterns are desig-
nated as s (singlet), d (doublet), m (multiplet) and br (broad). Gel
permeation chromatography (GPC) measurements were carried out on
a Polymer Laboratories PL-GPC 220 instrument, using 1,2,4-trichloro-
benzene as the eluent (stabilized with 125 ppm BHT) at 135 °C. The
obtained molecular weight is relative to polystyrene standards. UVꢀvis
absorption spectra were taken by a Shimadzu UV-2401PC spectro-
photometer. For the measurement of thin films, polymers were spun-
coat onto precleaned glass slides from ∼10 mg/mL polymer solutions in
chlorobenzene. Cyclic voltammetry measurements were performed on a
HOMO ¼ ꢀðEox þ 4:8Þ ðeVÞ
LUMO ¼ ꢀðEred þ 4:8Þ ðeVÞ
Polymer Solar Cell Fabrication and Testing. Glass substrates
coated with patterned tin-doped indium oxide (ITO) were purchased
from Thin Film Devices, Inc. Prior to use, the substrates were cleaned via
9152
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
ultrasonication in acetone, deionized water, and then 2-proponal for
20 min each. The substrates were dried under a stream of nitrogen and
then treated with UV-ozone for over 30 min. A filtered dispersion of
PEDOT:PSS in water (Baytron PH500) was then spun-cast onto clean
ITO substrates at 4000 rpm for 60 s and then baked at 140 °C for 10 min
to give a thin film with a thickness of 40 nm. Blends of polymer and
PCBM (1:1 w/w, 10 mg/mL for polymers) were dissolved in corre-
sponding solvents with heating at 120 °C for 6 h. All the solutions were
filtered through a 1.0 μm poly(tetrafluoroethylene) (PTFE) filter and
spun-cast at optimized rpm for 60 s onto the PEDOT:PSS layer. The
substrates were then dried at room temperature in the glovebox under a
nitrogen atmosphere for 12 h. The devices were finished for measure-
ment after thermal deposition of a 40 nm film of calcium and a 70 nm
aluminum film as the cathode at a pressure of ∼1 ꢁ 10ꢀ6 mbar. There
are 8 devices per substrate, with an active area of 12 mm2 per device. The
thicknesses of films were recorded by a profilometer (Alpha-Step 200,
Tencor Instruments), and AFM Images were taken using an Asylum
Research MFP3D atomic force microscope. Device characterization was
carried out under AM 1.5G irradiation with the intensity of 100 mW/cm2
(Oriel 91160, 300 W) calibrated by a NREL certified standard silicon
cell. Current densities versus potential (JꢀV) curves were recorded
with a Keithley 2400 digital source meter. EQE were detected under
monochromatic illumination (OrielCornerstone 260 1/4 m monochro-
mator equipped with Oriel 70613NS QTH lamp), and the calibration of
the incident light was performed with a monocrystalline silicon diode.
All fabrication steps after adding the PEDOT:PSS layer onto ITO
substrate, and characterizations were performed in gloveboxes under a
nitrogen atmosphere.
10235), NSF CAREER Award (DMR-0954280), and NSF Grant
(CHE-1058626). We acknowledge Mr. Nabil Kleinhenz for the
synthesis of 2-(2-ethylhexyl)thieno[3,4-b]thiophene and Prof.
Richard Jordan and Mr. Nathan Contrella of the University of
Chicago for GPC measurements. We also want to thank Dr. Shubin
Liu of Research Computing Center at UNC Chapel Hill for DFT
calculations.
’ REFERENCES
(1) Brabec, C. J. Sol. Energy Mater. Sol. Cells 2004, 83, 273.
(2) G€unes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007,
107, 1324.
(3) Thompson, B. C.; Frꢀechet, J. M. J. Angew. Chem., Int. Ed. 2008,
47, 58.
(4) Xiao, S.; Price, S. C.; Zhou, H.; You, W. ACS Symp. Ser. 2010,
1034, 71.
(5) Cheng, Y.-J.; Yang, S.-H.; Hsu, C.-S. Chem. Rev. 2009, 109, 5868.
(6) Coffin, R. C.; Peet, J.; Rogers, J.; Bazan, G. C. Nature Chem.
2009, 1, 657.
(7) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.;
Moses, D.; Leclerc, M.; Lee, K.; Heeger, A. J. Nature Photonics 2009,
3, 297.
(8) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.;
Beaujuge, P. M.; Frꢀechet, J. M. J. J. Am. Chem. Soc. 2010, 132, 7595.
(9) Price, S. C.; Stuart, A. C.; Yang, L.; Zhou, H.; You, W. J. Am.
Chem. Soc. 2011, 133, 4625.
(10) Zhou, H.; Yang, L.; Stuart, A. C.; Price, S. C.; Liu, S.; You, W.
Angew. Chem., Int. Ed. 2011, 50, 2995.
For mobility measurements, the hole-only devices in a configuration
of ITO/PEDOT:PSS (40 nm)/copolymerꢀPCBM/Pd (50 nm) were
fabricated. The experimental dark current densities J of polymer:PCBM
blends were measured when applied with voltage from 0 to 6 V. The
applied voltage V was corrected from the built-in voltage Vbi which was
taken as a compensation voltage Vbi = Voc + 0.05 V and the voltage drop
Vrs across the indium tin oxide/poly(3,4-ethylenedioxythiophene):poly-
(styrenesulfonic acid) (ITO/PEDOT:PSS) series resistance and contact
resistance, which is found to be around 35 Ω from a reference device
without the polymer layer. From the plots of J0.5 vs V, hole mobilities of
copolymers can be deduced from the equation
(11) Chu, T.-Y.; Lu, J.; Beauprꢀe, S.; Zhang, Y.; Pouliot, J.-R.; Wakim,
S.; Zhou, J.; Leclerc, M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem. Soc. 2011,
133, 4250.
(12) Su, M.-S.; Kuo, C.-Y.; Yuan, M.-C.; Jeng, U. S.; Su, C.-J.; Wei,
K.-H. Adv. Mater. 2011, 23, 3315.
(13) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu,
L. Adv. Mater. 2010, 22, E135.
(14) Liang, Y.; Feng, D.; Wu, Y.; Tsai, S.-T.; Li, G.; Ray, C.; Yu, L.
J. Am. Chem. Soc. 2009, 131, 7792.
(15) Hou, J.; Chen, H.-Y.; Zhang, S.; Chen, R. I.; Yang, Y.; Wu, Y.; Li,
G. J. Am. Chem. Soc. 2009, 131, 15586.
(16) Liang, Y.; Yu, L. Acc. Chem. Res. 2010, 43, 1227.
(17) Huang, Y.; Huo, L.; Zhang, S.; Guo, X.; Han, C. C.; Li, Y.; Hou,
J. Chem. Commun. 2011, 47, 8904.
9
V2
L3
J ¼ εrε0μh
8
(18) Wakim, S.; Alem, S.; Li, Z.; Zhang, Y.; Tse, S.-C.; Lu, J.; Ding, J.;
Tao, Y. J. Mater. Chem. 2011, 21, 10920.
(19) Roncali, J. Macromol. Rapid Commun. 2007, 28, 1761.
(20) Price, S. C.; Stuart, A. C.; You, W. Macromolecules 2010,
43, 4609.
where ε0 is the permittivity of free space, εr is the dielectric constant of
the polymer which is assumed to be around 3 for the conjugated poly-
mers, μh is the hole mobility, V is the voltage drop across the device, and
L is the film thickness of active layer.
(21) Zhou, H.; Yang, L.; Stoneking, S.; You, W. ACS Appl. Mater.
Interfaces 2010, 2, 1377.
(22) Zhou, H.; Yang, L.; Liu, S.; You, W. Macromolecules 2010,
43, 10390.
(23) Coffin, R. C.; Peet, J.; Rogers, J.; Bazan, G. C. Nature Chem.
2009, 1, 657.
(24) Zhou, H.; Yang, L.; Xiao, S.; Liu, S.; You, W. Macromolecules
2010, 43, 811.
’ ASSOCIATED CONTENT
S
Supporting Information. AFM images of PBnDT-TT,
b
DFT data, mobility data, syntheses of 3-bromothiophene-4-
carboxylic acid and 3-amino-4-bromothiophene, and NMR spectra
of all compounds and polymers. This material is available free of
charge via the Internet at http://pubs.acs.org.
(25) Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793.
(26) Liang, Y.; Wu, Y.; Feng, D.; Tsai, S.-T.; Son, H.-J.; Li, G.; Yu, L.
J. Am. Chem. Soc. 2009, 131, 56.
’ AUTHOR INFORMATION
(27) Zhou, H.; Yang, L.; Price, S. C.; Knight, K. J.; You, W. Angew.
Chem., Int. Ed. 2010, 49, 7992.
(28) Kim, I. T.; Lee, J. H.; Lee, S. W. Bull. Korean Chem. Soc. 2007,
28, 2511.
Corresponding Author
*E-mail: wyou@email.unc.edu.
(29) Pan, H.; Li, Y.; Wu, Y.; Liu, P.; Ong, B. S.; Zhu, S.; Xu, G. J. Am.
Chem. Soc. 2007, 129, 4112.
(30) Pan, H.; Wu, Y.; Li, Y.; Liu, P.; Ong, B. S.; Zhu, S.; Xu, G. Adv.
Funct. Mater. 2007, 17, 3574.
’ ACKNOWLEDGMENT
This work was generously supported by a DuPont Young
Professor Award, Office of Naval Research (Grant N0001411-
9153
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154

Macromolecules
ARTICLE
€
(31) Allard, N.; Beauprꢀe, S.; Aich, B. R.; Najari, A.; Tao, Y.; Leclerc,
M. Macromolecules 2011, 44, 7184.
(32) Yao, Y.; Liang, Y.; Shrotriya, V.; Xiao, S.; Yu, L.; Yang, Y. Adv.
Mater. 2007, 19, 3979.
(33) Kolb, H. C.; Finn, M. G.; Sharpless, K. B. Angew. Chem., Int. Ed.
2001, 40, 2004.
(34) Evindar, G.; Batey, R. A. J. Org. Chem. 2006, 71, 1802.
(35) Zhai, L.; Pilston, R. L.; Zaiger, K. L.; Stokes, K. K.; McCullough,
R. D. Macromolecules 2002, 36, 61.
(36) Spagnolo, P.; Zanirato, P. J. Org. Chem. 1978, 43, 3539.
(37) Bures, J.; Martin, M.; Urpi, F.; Vilarrasa, J. J. Org. Chem. 2009,
74, 2203.
(38) Bao, Z.; Chan, W. K.; Yu, L. J. Am. Chem. Soc. 1995, 117, 12426.
(39) Carsten, B.; He, F.; Son, H. J.; Xu, T.; Yu, L. Chem. Rev. 2011,
111, 1493.
(40) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C.
Adv. Mater. 2011, 23, 2367.
(41) Szarko, J. M.; Guo, J.; Liang, Y.; Lee, B.; Rolczynski, B. S.;
Strzalka, J.; Xu, T.; Loser, S.; Marks, T. J.; Yu, L.; Chen, L. X. Adv. Mater.
2010, 22, 5468.
(42) Kleinhenz, N.; Yang, L.; Zhou, H.; Price, S. C.; You, W.
Macromolecules 2011, 44, 872.
9154
dx.doi.org/10.1021/ma201835h |Macromolecules 2011, 44, 9146–9154