Conjugated polymers based on C, Si and N-bridged dithiophene and thienopyrroledione units: Synthesis, field-effect transistors and bulk heterojunction polymer solar cells

Article abstract of DOI:10.1039/c0jm03927f

A series of low band-gap conjugated polymers (PDTC, PDTSi and PDTP) containing electron-rich C-, Si-, and N-bridged bithiophene and electron-deficient thienopyrroledione units were synthesized via Stille coupling polymerization. All these polymers possess a low-lying energy level for the highest occupied molecular orbital (HOMO) (as low as -5.44 eV). As a result, photovoltaic devices derived from these polymers show high open circuit voltage (V_oc as high as 0.91 V). These rigid polymers also possess respectable hole mobilities of 1.50 × 10^-3, 6.0 × 10 ^-4, and 3.9 × 10^-4 cm² V^-1 s^-1 for PDTC, PDTSi, and PDTP, respectively. The combined high V _oc and good hole mobility enable bulk hetero-junction photovoltaic cells to be fabricated with relatively high power conversion efficiency (PCE as high as 3.74% for the PDTC-based device). The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c0jm03927f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2011, 21, 3895
www.rsc.org/materials
PAPER
Conjugated polymers based on C, Si and N-bridged dithiophene and
thienopyrroledione units: synthesis, field-effect transistors and bulk
heterojunction polymer solar cells
Yong Zhang, Jingyu Zou, Hin-Lap Yip, Ying Sun, Josh A. Davies, Kung-Shih Chen, Orb Acton
*
and Alex K.-Y. Jen
Received 15th November 2010, Accepted 4th December 2010
DOI: 10.1039/c0jm03927f
A series of low band-gap conjugated polymers (PDTC, PDTSi and PDTP) containing electron-rich
C-, Si-, and N-bridged bithiophene and electron-deficient thienopyrroledione units were synthesized via
Stille coupling polymerization. All these polymers possess a low-lying energy level for the highest
occupied molecular orbital (HOMO) (as low as ꢀ5.44 eV). As a result, photovoltaic devices derived
from these polymers show high open circuit voltage (Voc as high as 0.91 V). These rigid polymers also
possess respectable hole mobilities of 1.50 ꢁ 10ꢀ3, 6.0 ꢁ 10ꢀ4, and 3.9 ꢁ 10ꢀ4 cm2 Vꢀ1 sꢀ1 for PDTC,
PDTSi, and PDTP, respectively. The combined high Voc and good hole mobility enable bulk hetero-
junction photovoltaic cells to be fabricated with relatively high power conversion efficiency (PCE as
high as 3.74% for the PDTC-based device).
corresponding electrodes can also affect Voc.18–20 As a result, it is
1. Introduction
very critical to develop polymers with small band-gap and good
hole mobility, while maintaining a low lying HOMO energy level
to ensure high Voc.
Bulk hetero-junction-based (BHJ) polymer photovoltaic cells
have been vigorously investigated due to their potential for high-
speed processing using roll-to-roll printing to meet the scalability
challenge for low-cost renewable energy.1–4 The most frequently
studied BHJ cells involve the blending of a conjugated polymer
(electron donor) and a fullerene derivative (electron acceptor) to
form a nanoscale interpenetrating network to serve as the active
layer. The morphology of this active layer can be optimized by
either solvent or thermal annealing using mixed solvents or
additives.5–10 For example, the regioregular poly(3-hexyl-
thiophene) (P3HT) based devices have been demonstrated to be
able to achieve ꢂ5% power conversion efficiency (PCE) using
these methods.9,11 However, the relatively large band-gap and
high HOMO energy level of P3HT limit the possibility of further
improving device performance. More recently, significant prog-
ress has been made in new material development to enable
devices with very high PCEs ($7%).12–17 In order to obtain higher
PCE, the short circuit current (Jsc), open circuit voltage (Voc),
and fill factor (FF) of the device need to be optimized simulta-
neously.1 The Jsc is correlated to the absorption of the active
layer, therefore, better matched absorption with the solar spec-
trum will potentially give higher Jsc. Voc is mainly determined by
the energy offset between the HOMO of the polymer donor and
the lowest unoccupied molecular orbital (LUMO) of the
fullerene derivatives, although the interface between the
Compared to conjugated homopolymers (such as P3HT),
donor–acceptor (D–A) conjugated polymers, with electron-rich
and electron-deficient units alternating on the polymer back-
bone, are proven to be an effective approach to make low band-
gap polymers. The band-gap and energy level of these conjugated
polymers can be easily tuned by using suitable donor and
acceptor units, such as fluorene, silafluorene, carbazole, benzo-
dithiophene, benzothiadiazole, quinoxaline, and thienopyrazine,
etc., to achieve PCE of higher than 5%.7,8,12–17
Among these conjugating moieties, a dithiophene unit bridged
with a carbon (C) or a silicon (Si) atom has been incorporated
into alternating copolymers with an electron-deficient benzo-
thiadiazole unit to make efficient polymer BHJ cells.10,21 Origi-
nally, a low PCE of ca. 2.7% was achieved for devices containing
the C-bridged dithiophene-based polymer donor (PCPDTBT)
and the PC61BM acceptor.21 By optimizing the morphology of
the PCPDTBT/PC71BM blend with a small amount of alkyle-
nedithiol additive, the PCE went up to 5.5%.10 However, other
conjugated polymers with the same donor unit can only get low
PCE of 1–2%.22,23 This shows that the morphologies of these
types of polymers are quite sensitive to the solvents or additives
used for device processing. In addition to PCPDTBT, its Si
analog, PSBTBT, has also been reported by Hou et al.17 and
Brabec et al.24 as a promising polymer donor for PSCs. By
changing the bridging atom from C to Si, the PCE of the devices
can improve to 5.1% without adding any additives during the
device fabrication.17 This is possible because the longer C–Si
Department of Materials Science and Engineering, University of
Washington, Seattle, Washington, 98195, USA. E-mail: ajen@
u.washington.edu; Fax: +1 206 543 3100; Tel: +1 206 543 2626
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3895–3902 | 3895

View Article Online
ꢀ
bond (ꢂ0.3 A) on PSBTBT allows for better inter-chain polymer
(2-ethylhexyl)-dithieno[3,2-b:20,30-d]silole (3), 3,30-dibromo-2,20-
bithiophene, N-[1-(20-ethylhexyl)-3-ethylheptanyl]-dithieno[3,2-
b:20,30-d]pyrrole (5) and 5-octylthieno[3,4-c]pyrrole-4,6-dione
were prepared by following literature procedures.25,31,32 UV-Vis
spectra were obtained using a Perkin-Elmer Lambda-9 spectro-
packing to improve hole mobility. Compared to the C- and Si-
bridged dithiophene, dithieno[3,2-b:20,30-d]pyrrole (DTP), the N-
bridged dithiophene, possesses stronger electron-donating ability
and has been used as an electron-rich donor for D–A conjugated
polymers. For example, DTP-based polymers such as
PDTPDTBT and PDTPDTDPP containing either bis(2-thienyl)-
2,1,3-benzothiadiazole-50,500-diyl or a 3,6-dithiophen-2-yl-2,5-
dihydropyrrolo[3,4-c]pyrrole-1,4-dione (DPP) as the electron-
deficient unit have been reported by Zhou et al. that show broad
absorption almost reaching 1100 nm. However, BHJ cells
derived from these polymers only show relatively low PCE of
2.18% and 2.71%.25,26 This is due to the low Voc (0.37–0.62 V) of
these devices (as a result of high HOMO energy level of these
polymers) that limits their performance. Therefore, it is critical to
develop polymers with optimal HOMO energy levels and charge-
transporting properties to improve device performance.
1
photometer. The H NMR spectra were collected on a Bruker
AV 300 or 500 spectrometer operating at 300 or 500 MHz in
deuterated chloroform solution with TMS as reference. Cyclic
voltammetry of polymer films was conducted in acetonitrile with
0.1 M of tetrabutylammonium hexafluorophosphate using a scan
rate of 100 mV sꢀ1. ITO, Ag/AgCl and Pt mesh were used as
working electrode, reference electrode and counter electrode,
respectively. Differential scanning calorimetry (DSC) was per-
formed using a DSC2010 (TA instruments) under a heating rate
of 10 C minꢀ1 and a nitrogen flow of 50 mL minꢀ1. GPC and
ꢃ
preparative GPC were performed on a Waters 410 differential
refractometer with two columns connected in series with a THF
(the mobile phase) flowing rate of 1 mL minꢀ1. Monodisperse
polystyrene samples were used as the standard for the determi-
nation of molecular weight. AFM images under tapping mode
were taken on a Veeco multimode AFM with a Nanoscope III
controller.
In this paper, we present a series of new D–A alternating
polymers based on C-, Si- and N-bridged dithiophene as the
electron-rich donor and thienopyrroledione (TPD) as the elec-
tron-deficient acceptor. Recently, we and others have reported
devices based on polymer containing thienopyrroledione (TPD)
and benzodithiophene units that show high Voc (as high as
0.89 V).27–30 This inspires us to further explore the possibility of
using TPD to copolymerize with electron-rich bridged dithio-
phenes to improve polymer properties. These polymers were
Device fabrication and characterization
To fabricate conventional configuration solar cells, ITO-coated
glass substrates (15 U sqꢀ1) were first cleaned with detergent, de-
ionized water, acetone, and isopropyl alcohol. A thin layer (ca.
40 nm) of PEDOT:PSS (Baytronꢀ P VP AI 4083, filtered at 0.45
mm) was first spin-coated on the pre-cleaned ITO-coated glass
synthesized via
a Stille coupling polymerization reaction
(Scheme 1 and 2). The optical, electrochemical, charge-trans-
porting, and photovoltaic properties have been investigated. The
results from electrochemical studies showed that these polymers
have relatively low HOMO energy levels of ca. ꢀ5.40 eV,
therefore, Voc of the photovoltaic cells derived from these
materials is as high as 0.91 V, which is among the highest
obtained for bridged dithiophene-based polymer photovoltaic
devices.
ꢃ
substrates at 5000 rpm and baked at 140 C for 10 min under
ambient conditions. The substrates were then transferred into an
argon-filled glove-box. Subsequently, the polymer:PC71BM
active layer (ca. 90 nm) was spin-coated onto the PEDOT:PSS
layer at 900 rpm from a homogeneous blend solution. The
solution was prepared by dissolving the polymer at a blend
weight ratio of 1 : 2 in o-dichlorobenzene (or o-dichlorobenzene
with 2 vol% of 1-chloronaphthalene) and filtered with a 0.2 mm
PTFE filter. At the final stage, the substrates were pumped under
high vacuum (<2 ꢁ 10ꢀ6 Torr), and calcium (0.8 nm) topped with
aluminium (100 nm) was thermally evaporated onto the active
layer through a shadow mask to define the active area of the
devices. The device pattern was a 2 mm diameter circle, which
had nominal device area of 3.14 ꢁ 10ꢀ2 cm2. The un-encapsu-
lated solar cells were tested under ambient conditions using
a Keithley 2400 SMU and an Oriel Xenon lamp (450 W) with an
AM1.5 filter. A mask was used to define the device illumination
area of 10.08 mm2 to minimize photocurrent generation from the
edge of the electrodes. The light intensity was calibrated to
100 mW cmꢀ2 using a calibrated silicon solar cell with a KG5
filter, which had been previously standardized at the National
Renewable Energy Laboratory.
2. Experimental section
Materials and general characterization method
All reagents were purchased from commercial sources without
further purification. 4H-Cyclopenta[2,1-b:3,4-b0]dithiophene,
4,4-di(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene (1),
3,30-dibromo-5,50-bis(trimethylsilyl)-2,20-bithiophene, 4,40-bis-
Synthesis
Compound 2. To a solution of compound 1 (230 mg, 0.57
mmol) in THF (10 mL) was added n-BuLi (1 mL, 2 M in hexane)
at ꢀ78 ꢃC. The resulting mixture was kept at ꢀ78 ꢃC for 30 min
ꢃ
Scheme 1 Synthetic route for monomers.
then at room temperature for 30 min. After cooling to ꢀ78 C,
3896 | J. Mater. Chem., 2011, 21, 3895–3902
This journal is ª The Royal Society of Chemistry 2011

View Article Online
1
a similar procedure to that of PDTC (120 mg, 55%). H NMR
trimethyltin chloride (1.5 mL, 1 M in hexane) was added in one
portion. The mixture was poured into water after stirring at
room temperature overnight and extracted with hexane. The
organic phase was dried over anhydrous Na2SO4. After
removing the solvent under vacuum, a black liquid was achieved
and used in the next step without further purification (280 mg,
67%). 1H NMR (300 MHz, CDCl3, d) 6.96 (t, 2H), 1.86 (m, 4H),
1.02–0.93 (m, 18H), 0.76 (t, J ¼ 6.5 Hz, 6H), 0.60 (t, J ¼ 7.3 Hz,
6H), 0.37 (t, 18H). HRMS (ESI, m/z): [M + H]+ calcd for
C31H54S2Sn2, 730.1711; found, 730.1699.
(500 MHz, CDCl3, d) 8.12–7.98 (m, 2H), 3.73 (br, 2H), 2.05 (br,
2H), 1.74–0.80 (m, 47H). Mn ¼ 13.6 kDa, Mw ¼ 19.4 kDa, PDI ¼
1.43.
Synthesis of PDTP. PDTP was synthesized with monomer 6
(240 mg, 0.32 mmol) and 7 (127 mg, 0.30 mmol) by following
1
a similar procedure to that of PDTC (110 mg, 54%). H NMR
(500 MHz, CDCl3, d) 8.39 (br, 2H), 4.64 (br, 1H), 3.74 (br, 2H),
2.24 (br, 2H), 1.74–0.76 (m, 47H). Mn ¼ 10.6 kDa, Mw ¼ 13.7
kDa, PDI ¼ 1.29.
Compound 4. Compound 4 was prepared by following the
similar procedure for making compound 2 from compound 3 in
a yield of 77%. 1H NMR (300 MHz, CDCl3, d) 7.09 (s, 2H), 1.43
(m, 4H), 1.31–1.16 (m, 18H), 0.84–0.76 (m, 12H), 0.39 (s, 18H).
3. Results and discussion
Synthesis
Compound 6. To a solution of compound 5 (300 mg, 0.72
mmol) in THF (15 mL) was added t-BuLi (1 mL, 1.7 M in
hexane) at 0 ꢃC. The resulting mixture was kept at 0 ꢃC for 1 h.
Trimethyltin chloride (2 mL, 1 M in hexane) was added in one
portion. The mixture was poured into water after stirring at
room temperature overnight and extracted with hexane. The
organic phase was dried over anhydrous Na2SO4. After
removing the solvent under vacuum, a yellow liquid was achieved
and used in the next step without further purification (400 mg,
75%). 1H NMR (300 MHz, CDCl3, d) 7.00 (s, 2H), 4.48 (m, 1H),
2.12 (m, 2H), 1.63 (m, 2H), 1.22–0.60 (m, 30H), 0.38 (s, 18H).
Scheme 1 shows the synthesis of the dithiophene-based mono-
mers. Compound 1 was prepared from the alkylation of 4H-
cyclopenta[2,1-b:3,4-b0]dithiophene using 2-ethylhexylbromide
in the presence of DMSO and aqueous KOH.31 Direct lithiatiꢃon
of compound 1 to remove a-H of the thiophene unit at ꢀ78 C
and subsequent quenching with trimethyltin chloride gave the
corresponding di(trimethyl)stannyl compound 2 in 67% yield.
The synthesis of compound 3 involves the double halogen–
lithium exchange of bromine in 3,30-dibromo-5,50-bis-
(trimethylsilyl)-2,20-bithiophene using n-BuLi and subsequent
ring closure by adding di(2-ethylhexyl)dichlorosilane.17 Then,
a similar procedure used to make compound 2 was employed to
give compound 4 in 77% yield. Compound 5 was prepared from
4,40-dibromo-2,20-bithiophene by the ring cyclization using
a mixture of Pd2(dba)3 and BINAP as catalyst.25 After careful
1,3-Dibromo-5-octylthieno[3,4-c]pyrrole-4,6-dione (7). To
a solution of 5-octylthieno[3,4-c]pyrrole-4,6-dione (0.5 g, 1.9
mmol) in trifluoroacetic acid (5 mL) and conc. H2SO4 (2 mL) was
added NBS (1.1 g, 6.2 mmol) in portions. The mixture was stirred
at room temperature overnight. Then, water (50 mL) was added
and the mixture was extracted with dichloromethane twice. The
organic phase was dried over Na2SO4. After removing solvent
under vacuum, the crude product was purified by silica column
chromatography to give the title compound as a white solid in
75% yield. 1H NMR (CDCl3, ppm): 3.61 (t, J ¼ 7.2 Hz, 2H), 1.65
(m, 2H), 1.31 (m, 10H), 0.89 (t, J1 ¼ 6.2 Hz, J2 ¼ 6.9 Hz, 3H).
HRMS (ESI) (M+, C14H17Br2NO2S): calcd, 420.9347; found,
420.9341.
purification, the corresponding stannyl compound
6 was
achieved by direct lithiation using t-BuLi and quenching with
trimethyltin chloride in 75% yield. The further purification of
compounds 2, 4 and 6 through silica (or alumina) column
chromatography will lead to significant decomposition to the
compounds 1, 3 and 5, respectively. Compound 7 was made from
multiple steps starting from 3,4-dibromothiophene using the
literature reported method.27,32
The polymers were synthesized by reacting the corresponding
bis-stannyl compounds (2, 4 and 6) with 7 under the Stille
polymerization conditions using toluene–DMF as mixed
solvents and Pd(PPh3)4 as catalyst under microwave heating
(Scheme 2). The resulting crude polymers were purified by
Soxhlet extraction with methanol and acetone, successively, to
remove low molecular weight oligomers and residual catalyst.
The collected polymers were further purified by preparative GPC
using THF as the moving phase. The final polymers were
collected by precipitating into methanol, filtered and dried under
vacuum. The polymers show good solubility in common organic
solvents such as chloroform, THF, chlorobenzene, and dichlo-
robenzene. The molecular weights of the polymers were deter-
mined by GPC against the polystyrene standard (Table 1). The
number-average molecular weights (Mn) of PDTC, PDTSi, and
PDTP are 16.0 kDa, 13.6 kDa and 10.6 kDa, respectively, with
polydispersity indexes (PDI) of 1.26, 1.43 and 1.29. Thermal
properties of these polymers were evaluated with differentꢃial
scanning calorimetry (DSC). The samples were heated at 10 C
minꢀ1 under nitrogen. The DSC scan of PDTC showed a clear
Synthesis of PDTC. In a 25 mL round bottom flask, monomer
2 (254 mg, 0.35 mmol) and 7 (134 mg, 0.32 mmol) were charged
with a condenser under N2 protection. Then, dry toluene (3 mL),
DMF (0.5 mL) and Pd(PPh3)4 (7 mg) were added into the flask
consequently. The resulting mixture was degassed twice and
heated in a microwave reactor with 100 ꢃC for 10 min, 120 ꢃC for
50 min. After cooling to room temperature, the mixture was
poured into methanol. The precipitate was collected and washed
with acetone for 24 h with a Soxhlet apparatus. Further purifi-
cation was performed in a preparative GPC using THF as the
mobile phase. The recovered black solid from THF solution was
1
collected and dried overnight under vacuum (130 mg, 62%). H
NMR (500 MHz, CDCl3, d) 8.07–7.99 (m, 2H), 3.75 (br, 2H),
2.04 (br, 3H), 1.76 (br, 2H), 1.41–1.29 (m, 10H), 1.05–0.69 (m,
34H). Mn ¼ 16.0 kDa, Mw ¼ 20.2 kDa, PDI ¼ 1.26.
Synthesis of PDTSi. PDTSi was synthesized with monomer 4
(280 mg, 0.37 mmol) and 7 (136 mg, 0.32 mmol) by following
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3895–3902 | 3897

View Article Online
Scheme 2 Synthesis of polymers PDTC, PDTSi and PDTP.
glass transition at ꢂ180 ꢃC, however, PDTSi and PTDP did not
Fig. 1 UV-Vis spectra of polymers PDTC, PDTSi and PDTP in (a)
ꢃ
chloroform solution and in (b) film states.
show any transitions between 20 and 350 C.
Optical properties
PDTSi and PDTP, respectively by calculating from the absorp-
tion edge of these films (Table 1).
Optical properties of these polymers were investigated in their
chloroform solutions and in thin films. Fig. 1a shows the
absorption spectra of all three polymers in chloroform. It can be
seen from the spectra that the absorption maximum changes
from 653 nm in PDTC with a high-energy shoulder at 604 nm, to
608 nm in PDTSi with a low-energy shoulder at 660 nm, and to
645 nm in PDTP with a high-energy shoulder at 598 nm. The
absorption edge of PDTC and PDTSi are similar at ꢂ723 nm,
while PDTP has a lower-energy absorption edge at 747 nm. In
the thin film state (Fig. 1b), the predominant peaks move to 671,
665, and 686 nm in PDTC, PDTSi and PDTP, respectively. The
red-shift of absorption in the thin film state implies a slightly
increased p–p stacking between polymer chains. On the other
hand, the slight red-shift of the lmax of PDTP is consistent with
the stronger electron-donating properties of the N atom
compared to either C or Si atoms.25 The absorption edges of
PDTC, PDTSi and PDTP in their film state are 741, 734 and 778
nm, respectively. The corresponding optical bandgaps of these
polymers were estimated to be 1.67, 1.70 and 1.59 eV for PDTC,
Electrochemical properties
To investigate the oxidative and reductive behaviors of these
polymers, cyclic voltammograms (CVs) were performed for thin
films prepared from these polymers on ITO with a Pt counter
electrode and a Ag/Ag+ reference electrode in 0.1 M tetrabuty-
lammonium hexafluorophosphate in MeCN at a scan rate of
100 mV sꢀ1. The HOMO and LUMO energy levels were calcu-
lated using the equations33
HOMO ¼ ꢀ[Eox ꢀ Eferrocene + 4.80] eV
LUMO ¼ ꢀ[Ered ꢀ Eferrocene + 4.80] eV
where Eox and Ered are the onset of the oxidation and reduction
potential, respectively. Eferrocene is the onset of the oxidation
Table 1 Molecular weight, PDI, thermal, optical and electrochemical data of polymers PDTC, PDTSi and PDTP
UV-Vis absorption
Cyclic voltammetry
solution
film
a
Polymer
Mn/kg molꢀ1
PDI
lmax/nm
lonset/nm
lmax/nm
lonset/nm
HOMO/eV
LUMO/eV
Band-gap
PDTC
PDTSi
PDTP
16.0
13.6
10.6
1.26
1.43
1.29
653; 604
660; 608
645; 598
724
723
747
671; 616
665; 609
686; 628
741
734
778
ꢀ5.43
ꢀ5.44
ꢀ5.16
ꢀ3.76b/ꢀ3.25c
ꢀ3.74b/ꢀ3.17c
ꢀ3.57b/ꢀ3.08c
1.67d/2.18e
1.70d/2.27e
1.59d/2.08e
Determined by GPC against polystyrene standard. b Calculated from the optical band-gap. c Measured from cyclic voltammetry. d Optical band-gap.
Electrochemical band-gap.
a
e
3898 | J. Mater. Chem., 2011, 21, 3895–3902
This journal is ª The Royal Society of Chemistry 2011

View Article Online
compounds were calculated by Density Functional Theory
(DFT) at the B3LYP/6-31G(d) level. The HOMO and LUMO
wave functions of the trimer model compounds are shown in
Fig. 3. It is noted that all side-chain substituents were replaced
with ethyl groups to save calculation time since this simplifica-
tion only has a minimal effect on the electronic properties. The
calculation for HOMO showed nice delocalization along the
whole conjugated backbone affected by both donor and acceptor
parts. Similarly, the calculated density of states for LUMO also
showed a nice distribution across the whole conjugated back-
bone, but with more effect on the acceptor part. In addition, the
effects of the bridging atom in the dithiophene unit are quite
discrete for both the HOMO and LUMO electron density
distributions. The densities of the respective HOMO wave
functions are distributed across the fused hetero-aromatic
backbones with less density on the electron-accepting substitu-
ents. For the LUMO, electron densities are localized to a greater
degree on the respective electron acceptor units and at the center
of the fused aromatic backbone. A final observation about the
phases of the atomic contributions at the double bonds to both
frontier MOs is warranted: in the HOMO wave function, the
carbon–carbon double bond units are p-bonding with alter-
nating phase with respect to adjacent double bonds. In the
LUMO, however, the former carbon–carbon double bond units
are p* and have the same phase as their neighbors. These results
show that both the donor and acceptor in these D–A polymers
can significantly affect the energy levels.
Fig. 2 CV curves of polymers.
potential of ferrocene. Fig. 2 shows the CV curves of the poly-
mers and the detailed data are summarized in Table 1. As
a comparison, the LUMO levels of these polymers were also
calculated from the HOMO levels measured from CV and the
optically measured bandgaps. The HOMO energy levels
measured by CV for PDTC, PDTSi, and PDTP are ꢀ5.43, ꢀ5.44
and ꢀ5.16 eV, respectively, which are 0.20–0.60 eV lower than
those reported for C- or Si-bridged bithiophene-based poly-
mers.15,17 The LUMO energy levels measured by CV are ꢀ3.25,
ꢀ3.17 and ꢀ3.08 eV, respectively. However, the calculated
LUMO energy levels (the difference between HOMO and Egopt
)
are ꢀ3.76, ꢀ3.74 and ꢀ3.57 eV, respectively, thus the electro-
chemical band-gap (Egec) is larger than the optical band-gap
Egopt. Similar phenomena have been reported in C- and Si-
bridged bithiophene-based polymers.17,22–26
OFET properties
Carrier mobility also plays an important role in the performance
of BHJ PSCs and a carrier mobility of larger than 10ꢀ4 cm2 Vꢀ1
sꢀ1 is needed for the polymer donor. To measure hole mobility,
top contact organic field-effect transistors (OFET) coated with
these polymers were fabricated using gold as the source and drain
electrodes. Interdigitated source (s) and drain (d) electrodes
(W ¼ 9000 mm, L ¼ 90 mm, W/L ¼ 100) were defined on top of
the polymer by evaporating a 50 nm thick gold film through
a shadow mask. The saturation field-effect mobility (m) was
calculated from the following equation:
Usually, the bandgap and HOMO/LUMO energy levels
originate from the molecular orbital hybridization of donor and
acceptor energy levels in the donor–acceptor type polymers.
Therefore, electron-donating groups will reduce Eg by raising the
HOMO energy level in the polymer. In our case, the cyclo-
pentadithiophene donor unit in PDTC and dithienosilole donor
unit in PDTSi should have similar electron-donating abilities due
to their similar experimental bandgaps and energy levels (Fig. 2,
Table 1). However, in the PDTP polymer, the electron-donating
ability of N atom is much stronger than both C and Si atoms,
therefore, a lower bandgap and a higher HOMO energy level for
this polymer were found.25
Ids ¼ (W/2L)mCi(Vgs ꢀ Vth)2
where W and L are the channel width and length, respectively. Ci
is the capacitance of the insulating SiO2 layer per unit, Vgs and
Vth are the gate voltage and the threshold voltage, respectively.34
In BHJ photovoltaic cells, the proper energy offset between
LUMO of donor and acceptor is required to provide energetically
favorable condition for exciton dissociation.34 As shown in Table
1, the difference between the LUMO levels of three polymers and
PC71BM (ꢀ4.30 eV) is between 0.54 and 0.73 eV, which is suffi-
cient for achieving efficient exciton dissociation from the interface
of the polymer donor and PC71BM acceptor. In addition, Voc is
proportional to the energy offset between the HOMO energy level
of the polymer donor and the LUMO energy level of fullerene
derivatives. Therefore, in the polymer/PCBM system, a deeper
HOMO energy level will lead to higher Voc. As discussed above,
the deeper HOMO levels of these polymers compared to other
bithiophene-based polymers (ꢀ5.4 vs. ꢀ5.1 eV) give higher Voc in
BHJ photovoltaic cells and ultimately lead to improved efficiency.
To better understand the oxidative and reductive properties of
these polymers, the electronic properties of their trimer model
Fig. 3 The HOMO and LUMO wave functions of the trimer model of
polymers PDTC, PDTSi and PDTP.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3895–3902 | 3899

View Article Online
Fig. 4 shows the FET characteristics of the devices in p-
channel mode with PDTC, PDTSi and PDTP as the channel
materials. The hole-mobility of the polymer was calculated from
the transport characteristics of the FETs by plotting Ids with
respect to Vgs. The calculated hole-mobilities of PDTC, PDTSi
these polymers were first prepared from spin-coating their o-
dichlorobenzene (o-DCB) solutions and the corresponding
devices were fabricated and measured. The photovoltaic
performance including Voc, Jsc, FF, and PCE are summarized in
Table 2. The J–V curves are shown in Fig. 6a. For the PDTC
device, the result showed a PCE of 3.74%. Under the same
conditions, PCE of 1.18% and 0.91% were obtained for PDTSi
and PDTP, respectively. It is worth noting that these devices
show quite high Voc (up to 0.91 V in the PDTSi device), which is
ꢂ0.2–0.4 V higher than other devices made from C, Si and N-
bridged dithiophene polymers.10,17,22–26 The high Voc obtained in
these devices are the result from the low HOMO energy level of
these polymers (ꢀ5.43 eV for PDTC, ꢀ5.44 eV for PDTSi, and
ꢀ5.16 eV for PDTP).35
and PDTP are 1.50 ꢁ 10ꢀ3, 6.0 ꢁ 10ꢀ4 and 3.9 ꢁ 10ꢀ4 cm2 Vꢀ1 sꢀ1
,
respectively, with the on/off ratios of 104–105. These values are
within the desirable range for OPV applications and also indicate
that there exists significant intramolecular charge transfer or
p–p packing in these polymer films.
Photovoltaic properties
The photovoltaic properties of these polymers were investigated
using a standard device configuration of ITO/PEDOT:PSS/poly
mer:PC71BM (1 : 2)/Ca/Al and their performance were measured
under 100 mW cmꢀ2 AM 1.5 illumination. The active layers of
In general, the formation of a bicontinuous interpenetrating
network between polymer donor and PCBM acceptor is the most
promising architecture in BHJ cells. The morphology of the
active layer (polymer:PCBM blend film) is very critical for the
performance of BHJ cells. If there are large size domains and
significant phase separation in the active layer, it will reduce the
interfaces for efficient charge separation and a smooth film with
few domains will also increase the possibility for charge recom-
bination. As can be seen from Table 2, PDTSi and PDTP devices
fabricated from o-DCB solution show much poorer performance
compared to the PDTC device. The inferior device performance
may be due to the non-optimal morphology in these polymer/
PC71BM blend films.
To understand the morphological effect on the device perfor-
mance, the films of polymer blends were investigated by tapping-
mode atomic force microscopy (AFM) and these AFM
topography images are shown in Fig. 5. As shown in Fig. 5a, the
PDTC/PC71BM film shows a smoother surface and smaller phase
separation, which facilitates charge separation and results in
higher device performance. However, in the PDTSi/PC71BM and
PDTP/PC71BM blend films (Fig. 5b, c), significant phase sepa-
ration and large domains were observed. This leads to poorer
charge separation and increased charge carrier recombination.
The strong phase separation may be due to polymer self-aggre-
gation and/or incompatibility between the polymer and
PC71BM. Therefore, this results in poorer performance in PDTSi
and PTDP devices.
There are several ways to alter the interpenetrating nanoscale
morphology of a BHJ cell. Bazan et al. have demonstrated that
the addition of a small amount of solvent additives with high
boiling point into the conjugated polymer/PCBM solution
during processing can give a substantial increase of device
Table 2 Performance of photovoltaic devices under AM 1.5 simulated
illumination (100 mW cmꢀ2
)
Polymer:PC71BM (1 : 2)
Voc/V
Jsc/mA cmꢀ2
FF
PCE (%)
PDTCa
PDTCb
PDTSia
PDTSib
PDTPa
PDTPb
0.80
0.80
0.91
0.85
0.71
0.76
10.04
9.40
2.32
6.58
2.53
4.69
0.47
0.45
0.56
0.37
0.50
0.53
3.74
3.45
1.18
2.13
0.91
1.69
Fig. 4 The output at different gate voltage (Vgs) and transfer charac-
teristics in the saturation regime under constant source–drain voltage
(Vds ¼ ꢀ100 V) for PDTC (a), PDTSi (b) and PDTP (c).
a
b
Film from pristine DCB. Film from o-DCB : 2 vol% CN.
3900 | J. Mater. Chem., 2011, 21, 3895–3902
This journal is ª The Royal Society of Chemistry 2011

View Article Online
images. As shown from Fig. 5d, the addition of 2 vol% CN had
a minor effect on the morphology of the PDTC/PC71BM thin
film. However, the previously observed large size domains from
the PDTSi/PC71BM and PDTP/PC71BM thin films cast from
their o-DCB solutions (Fig. 5b, c) can no longer be detected. The
nanoscale morphologies of PDTSi/PC71BM and PDTP/PC71BM
films were substantially improved. This indicates that CN solvent
additive can promote the self-aggregation of PDTSi and PDTP
in a relatively fluid medium and also improves the compatibility
between polymer and PC71BM.37,38 As a result, relatively smooth
BHJ films of polymer/PC71BM were formed. To verify this point,
BHJ cells with the same device configuration as described above
were fabricated by spin-coating films from their o-DCB:2 vol%
CN solutions. The optimized volume of CN is 2 vol%. The J–V
curves of these devices are shown in Fig. 6b and summarized in
Table 2. As shown in Fig. 6b, there are significant improvements
in the device performance, especially for the PDTSi and PDTP
devices. The PCE of the PDTC device remained at 3.45% while
the PCE of the PDTSi device increased from 1.18% to 2.13% with
almost 3 times increase of its Jsc. However, Voc of the device
showed a slight decrease to 0.85 V. A similar decrease has also
been observed earlier10 which may be due to morphological
changes and interfacial interactions between the active layer and
the cathode. For the PDTP device, the PCE improved from
0.91% to 1.69% after the addition of CN.
Fig. 5 AFM topography images of polymer:PC71BM (1 : 2 wt) blend films
from o-DCB (a, b, c) and o-DCB : 2 vol% 1-chloronaphthalene (d, e, f).
performance due to significantly improved morphology. The
solvent additives play the role of slowing the aggregation of
conjugated polymer and PCBM so as to be able to avoid forming
the oversize polymer or PCBM aggregates, which inhibit charge
carrier transport in the BHJ film.36,37 Recently, 1-chloronaph-
thalene (CN) has been proved to be an alternative solvent
additive in improving the performance of bridged dithiophene-
based polymers.38,39 To take advantage of this effect, CN was
also used as the solvent additive to improve the morphology of
the polymer blends, especially for the PDTSi/PC71BM and
PDTC/PC71BM systems. The morphology of the polymer/
PC71BM films cast from their o-DCB solution with 2 vol% of CN
was studied by AFM. Fig. 5 shows the AFM topographical
The external quantum efficiency (EQE) of these devices was
measured to evaluate the photoresponse of the BHJ cells. Fig. 7
shows the EQE curves of the PDTC device fabricated from
o-DCB and the PDTSi and PTDP devices fabricated from o-
DCB:2 vol% CN solution. All devices exhibited efficient photo-
response between 350 and 750 nm. The EQE of the PDTC device
is more than 45% with the highest value of 53.5% at 570 nm.
However, the PDTSi and PDTP devices showed much lower
EQE values compared to the PDTC device with their maxima at
38.3% and 25.2% at 560 nm, respectively. The low EQE values
affect the low Jsc observed for PDTSi and PDTP devices
(Table 2). In addition, the low EQE in PDTSi and PDTP devices
was mainly ascribed to the lower hole mobility observed in these
two polymers because of non-ideal interpenetrating nanoscale
phase separation between the polymers and PC71BM. Since there
are many factors affecting the phase separation, such as the
Fig. 6 J–V curves of PDTC, PDTSi and PDTP devices: (a) film prepared
from pristine o-DCB solution and (b) film prepared from o-DCB:2
Fig. 7 External quantum efficiency curves of optimized PDTC, PDTSi
vol%1-chloronaphthalene under illumination of AM 1.5, 100 mW cmꢀ2
.
and PDTC devices.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 3895–3902 | 3901

View Article Online
11 L. J. A. Koster, V. D. Mihailetchi and P. W. M. Blom, Appl. Phys.
Lett., 2006, 88, 093511.
12 E. G. Wang, L. Wang, L. F. Lan, C. Luo, W. L. Zhuang, J. B. Peng
and Y. Cao, Appl. Phys. Lett., 2008, 92, 033307.
13 Y. Y. Liang, Y. Wu, D. Q. Feng, S. T. Tsai, H. J. Son, G. Li and
L. P. Yu, J. Am. Chem. Soc., 2009, 131, 56.
Flory–Huggins interaction, the tendency of the components to
aggregate, and kinetic constraints,10,40,41 further work will focus
on understanding the relationship between morphology and
performance and will be reported elsewhere.
14 Y. Y. Liang, D. Q. Feng, Y. Wu, S. T. Tsai, G. Li, C. Ray and
L. P. Yu, J. Am. Chem. Soc., 2009, 131, 7792.
4. Conclusion
15 R. C. Coffin, J. Peet, J. Rogers and G. C. Bazan, Nat. Chem., 2009, 1,
657.
16 J. H. Hou, H. Y. Chen, S. Q. Zhang, R. I. Chen, Y. Yang, Y. Wu and
G. Li, J. Am. Chem. Soc., 2009, 131, 15586.
17 J. H. Hou, H. Y. Chen, S. Q. Zhang, G. Li and Y. Yang, J. Am. Chem.
Soc., 2008, 130, 16144.
18 J. Luo, H. B. Wu, C. He, A. Y. Li, W. Yang and Y. Cao, Appl. Phys.
Lett., 2009, 95, 043301.
19 C. He, C. M. Zhong, H. B. Wu, R. Q. Yang, W. Yang, F. Huang,
G. C. Bazan and Y. Cao, J. Mater. Chem., 2010, 20, 2617.
20 V. D. Mihailetchi, P. W. M. Blom, J. C. Hummelen and
M. T. Rispens, J. Appl. Phys., 2003, 94, 6849.
21 D. Muhlbacher, M. Scharber, M. Morana, Z. G. Zhu, D. Waller,
R. Gaudiana and C. J. Brabec, Adv. Mater., 2006, 18, 2931.
Three new polymers, PDTC, PDTSi and PDTP, have been
designed and synthesized through the Stille coupling polymeri-
zation between C-, Si-, N-bridged dithiophene stannyl compound
and 1,3-dibromo-5-octylthieno[3,4-c]pyrrole-4,6-dione. The
change from C, Si to N atom in the bridged dithiophene induces
a red-shift in their thin film absorption spectra. The results from
electrochemical measurements showed that these polymers
possess lower HOMO levels (ꢀ5.44 eV for PDTSi, ꢀ5.43 eV for
PDTC and ꢀ5.16 eV for PDTP) compared to previously reported
analogs. These polymers also have good hole mobility (as high as
1.50 ꢁ 10ꢀ3) as measured by the FET technique. The photovoltaic
properties of these polymers were investigated using the device
configuration of ITO/PEDOT:PSS/polymer:PC71BM (1 : 2)/Ca/
Al. The highest achievable PCE for PTDC, PDTSi, and PDTP is
3.74%, 2.13%, and 1.69%, respectively. It is worth noting that the
Voc of these devices increased significantly (ꢂ0.2–0.4 V) due to the
lower HOMO energy level of these polymers compared to other
C-, Si-, N-bridged dithiophene-based polymers.
€
22 A. J. Moule, A. Tsami, T. W. Bunnagel, M. Forster,
N. M. Kronenberg, M. Scharber, M. Koppe, M. Morana,
C. J. Brabec, K. Meerholz and U. Scherf, Chem. Mater., 2008, 20,
4045.
23 I. H. Jung, H. Kim, M. J. Park, B. Kim, J. H. Park, E. Jeong,
H. J. Woo, S. Yoo and H. K. Shim, J. Polym. Sci., Part A: Polym.
Chem., 2010, 48, 1423.
24 M. C. Scharber, M. Koppe, J. Gao, F. Cordella, M. A. Loi, P. Denk,
M. Morana, H. J. Egelhaaf, K. Forberich, G. Dennler, R. Gaudiana,
D. Waller, Z. G. Zhu, X. B. Shi and C. J. Brabec, Adv. Mater., 2010,
22, 367.
25 E. J. Zhou, M. Nakamura, T. Nishizawa, Y. Zhang, Q. S. Wei,
K. Tajima, C. H. Yang and K. Hashimoto, Macromolecules, 2008,
41, 8302.
26 E. J. Zhou, Q. S. Wei, S. Yamakawa, Y. Zhang, K. Tajima,
C. H. Yang and K. Hashimoto, Macromolecules, 2010, 43, 821.
27 Y. Zhang, H. K. Steven, H. L. Yip, Y. Sun, O. Acton and
A. K. Y. Jen, Chem. Mater., 2010, 22, 2696.
28 Y. P. Zou, A. Najari, P. Berrouard, S. Beaupre, B. R. Aıch, Y. Tao
and M. Leclerc, J. Am. Chem. Soc., 2010, 132, 5330.
29 C. Piliego, T. W. Holcombe, J. D. Douglas, C. H. Woo, P. Beaujuge
and J. M. J. Frechet, J. Am. Chem. Soc., 2010, 132, 7595.
30 G. B. Zhang, Y. Y. Fu, Q. Zhang and Z. Y. Xie, Chem. Commun.,
2010, 46, 4997.
Acknowledgements
The authors are grateful for the support of the National Science
Foundation’s STC program under DMR-0120967, the AFOSR’s
‘‘Interface Engineering’’ Program (FA9550-09-1-0426), the DOE
‘‘Future Generation Photovoltaic Devices and Process’’ program
(DE-FC3608GO18024/A000), the Office of Naval Research
(N00014-08-1-1129) and the World Class University (WCU)
program through the National Research Foundation of Korea
under the Ministry of Education, Science and Technology (R31-
10035). A. K.-Y. Jen thanks the Boeing-Johnson Foundation for
financial support.
€
31 Z. G. Zhu, D. Waller, R. Gaudiana, M. Morana, D. Muhlbacher,
M. Scharber and C. J. Brabec, Macromolecules, 2007, 40, 1981.
32 C. B. Nielsen and T. Bjornholm, Org. Lett., 2004, 6, 3381.
33 J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler,
M. Porsch and J. Daub, Adv. Mater., 1995, 7, 551.
34 A. Zen, J. Pflaum, S. Hirschmann, W. Zhuang, F. Jaiser,
U. Asawapirom, J. P. Rabe, U. Scherf and D. Neher, Adv. Funct.
Mater., 2004, 14, 757.
Notes and references
1 C. J. Brabec, N. S. Sariciftci and C. J. Hummelen, Adv. Funct. Mater.,
2001, 11, 15.
2 F. C. Krebs, Sol. Energy Mater. Sol. Cells, 2009, 93, 394.
3 G. Dennler, M. C. Scharber and C. J. Brabec, Adv. Mater., 2009, 21,
1323.
€
35 M. C. Scharber, D. Muhlbacher, M. Koppe, P. Denk, C. Waldauf,
4 S. Gunes, H. S. Neugebauer and N. S. Sariciftci, Chem. Rev., 2007,
107, 1324.
5 N. S. Sariciftci, L. Smilowitz, A. J. Heeger and F. Wudl, Science,
1992, 258, 1474.
A. J. Heeger and C. J. Brabec, Adv. Mater., 2006, 18, 789.
36 J. K. Lee, W. L. Ma, C. J. Brabec, J. Yuen, J. S. Moon, J. Y. Kim,
K. Lee, G. C. Bazan and A. J. Heeger, J. Am. Chem. Soc., 2008,
130, 3619.
6 G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science,
1995, 270, 1789.
37 J. S. Moon, C. J. Takacs, S. Cho, R. C. Coffin, H. Kim, G. C. Bazan
and A. J. Heeger, Nano Lett., 2010, 10, 4005.
7 J. W. Chen and Y. Cao, Acc. Chem. Res., 2009, 42, 1709.
8 Y. J. Cheng, S. H. Yang and C. S. Hsu, Chem. Rev., 2009, 109, 5868.
9 G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery and
Y. Yang, Nat. Mater., 2005, 4, 864.
38 C. V. Hoven, X. D. Dang, R. C. Coffin, J. Peet, T.-Q. Nguyen and
G. C. Bazan, Adv. Mater., 2010, 22, E63.
39 R. C. Coffin, J. Peet, J. Rogers and G. C. Bazan, Nat. Chem., 2009, 1,
657.
10 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger
and G. C. Bazan, Nat. Mater., 2007, 6, 497.
40 H. Hoppe and N. S. Sariciftci, J. Mater. Chem., 2006, 16, 45.
41 X. N. Yang and J. Loos, Macromolecules, 2007, 40, 1353.
3902 | J. Mater. Chem., 2011, 21, 3895–3902
This journal is ª The Royal Society of Chemistry 2011