Low bandgap π-conjugated copolymers based on fused thiophenes and benzothiadiazole: Synthesis and structure-property relationship study

Authors

  • Shiming Zhang,

    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    2. Graduate University of Chinese Academy of Sciences, Beijing 100039, People's Republic of China
    Search for more papers by this author
  • Yunlong Guo,

    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    2. Graduate University of Chinese Academy of Sciences, Beijing 100039, People's Republic of China
    Search for more papers by this author
  • Haijun Fan,

    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    2. Graduate University of Chinese Academy of Sciences, Beijing 100039, People's Republic of China
    Search for more papers by this author
  • Yao Liu,

    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    2. Graduate University of Chinese Academy of Sciences, Beijing 100039, People's Republic of China
    Search for more papers by this author
  • Hsiang-Yu Chen,

    1. Department of Materials Science and Engineering, University of California, Los Angeles, California 90095
    Search for more papers by this author
  • Guanwen Yang,

    1. Department of Materials Science and Engineering, University of California, Los Angeles, California 90095
    Search for more papers by this author
  • Xiaowei Zhan,

    Corresponding author
    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    • Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    Search for more papers by this author
  • Yunqi Liu,

    Corresponding author
    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    • Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    Search for more papers by this author
  • Yongfang Li,

    1. Beijing National Laboratory for Molecular Sciences and CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, People's Republic of China
    Search for more papers by this author
  • Yang Yang

    1. Department of Materials Science and Engineering, University of California, Los Angeles, California 90095
    Search for more papers by this author

Abstract

A series of low bandgap conjugated polymers consisting of benzothiadiazole alternating with dithienothiophene (DTT) or dithienopyrrole (DTP) unit with or without 3-alkylthiophene bridge have been synthesized. Effect of the fused rings and 3-alkylthiophene bridge on the thermal, optical, electrochemical, charge transport, and photovoltaic properties of these polymers have been investigated. These polymers show broad absorption extending from 300 to 1000 nm with optical bandgaps as low as 1.2 eV; the details of which can be varied either by incorporating 3-alkylthiophene bridge or by replacing DTT with DTP. The LUMO levels (−2.9 to −3.3 eV) are essentially unaffected by the specific choice of donor moiety, whereas the HOMO levels (−4.6 to −5.6 eV) are more sensitive to the choice of donor. The DTT and DTP polymers with 3-alkylthiophene bridge were found to exhibit hole mobilities of 8 × 10−5 and 3 × 10−2 cm2 V−1 s−1, respectively, in top-contact organic field-effect transistors. Power conversion efficiencies in the range 0.17–0.43% were obtained under simulated AM 1.5, 100 mW cm−2 irradiation for polymer solar cells using the DTT and DTP-based polymers with 3-alkylthiophene bridge as donor and fullerene derivatives as acceptor. © 2009 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 47: 5498–5508, 2009

INTRODUCTION

Recently, there has been a considerable interest in low bandgap conjugated polymers, specifically those with bandgaps below 1.5 eV.1 Such materials can provide unique electronic and optical properties, such as near-infrared (NIR) absorption, NIR emission, and solid-state charge transport.2 The use of visible and NIR absorbing polymers could improve power conversion efficiency (PCE) of polymer solar cells (PSCs). NIR emitters can find applications in sensor, detector, and telecommunication technologies.3 On the other hand, high mobility of organic semiconductors is important for electronic devices, especially for organic field-effect transistors (OFETs) and PSCs. However, there have been only a few reports on low bandgap polymers with high mobility.2–5 One way to design low bandgap polymers is to alternate donor (D) and acceptor (A) units in the conjugated backbone of the polymers. This strategy can manipulate the electronic structure (HOMO/LUMO levels) through the partial intramolecular charge transfer (ICT) happening in the D-A systems.6, 7

Molecules containing fused-ring systems intend to maximize the π-orbital overlap by restricting intramolecular rotation in the oligomer and possibly to induce face-to-face π-stacking, facilitating charge transport through intermolecular hopping. On the other hand, the fused aromatic rings can make the polymer backbone more rigid and coplanar, therefore enhancing effective π-conjugation, lowering band gap and extending absorption. Thus, fused thiophenes, such as thienothiophene,8, 9 benzodithiophene,10 dithienocyclopentadiene (DTC),5, 11 dithienosilole (DTS),12, 13 and tetrathienoacene,14 are introduced into π-conjugated polymeric systems to obtain good performance in OFETs and PSCs.

Dithieno[3,2-b:2′,3′-d]thiophene (DTT) is sulfur rich and electron rich species, and serves as an important building block of a wide variety of materials for electronic and optical applications, such as electroluminescence, two-photon absorption, nonlinear optics, photochromism, OFETs, and PSCs.15 A few groups copolymerized DTT with alkyl-substituted thiophene and used the copolymers to fabricate OFETs16 and PSCs.17 Copolymers of DTT with fluorene18 or p-bis(cyanovinyl)phenylene19 emit intense fluorescence. Recently, we also reported synthesis of perylene diimide-DTT20, 21 and porphyrin-DTT22 copolymers and DTT homopolymers23 and their applications in PSCs and OFETs.

Dithieno[3,2-b:2′,3′-d]pyrroles (DTPs) have recently been incorporated into a range of conjugated polymers2, 24–29; compared with DTTs they are more electron rich and also offer the possibility for tuning of polymer solubility and ordering through the N substituents without disrupting the polymer backbone.2 OFETs2, 26, 27 and PSCs28, 29 based on DTP copolymers exhibited good performance.

2,1,3-Benzothiadiazole is a widely used acceptor for synthesis of D-A polymers. For example, copolymers of benzothiadiazole with fluorene,30, 31 silafluorene,32, 33 carbazole,34, 35 DTS,13, 36 DTC,11, 37 and DTP29 were synthesized and applied to PSCs yielding PCEs in the range of 0.18–5.4%. On the other hand, a copolymer of benzothiadiazole with DTC showed a hole mobility as high as 0.17 cm2 V−1 s−1.5

In this article, we report synthesis of a series of low bandgap π-conjugated copolymers of benzothiadiazole with DTT and DTP (1a-b, 2a-b, Fig. 1), some of which show strong NIR absorption and/or high mobility. We compare the effects of the fused thiophenes on the electronic properties of the polymers. We also compare the performance of PSCs and OFETs based on these polymers.

Figure 1.

Chemical structures of the fused thiophenes-benzothiadiazole copolymers discussed in this article.

RESULTS AND DISCUSSION

Synthesis and Characterization

The synthetic routes to the monomers and polymers are outlined in Scheme 1. The Stille coupling reaction between 5-tri-n-butylstannyl-3-n-dodecylthiophene and 4,7-dibromo-2,1,3-benzothiadiazole in the presence of a catalytic amount of Pd(PPh3)4 afforded the intermediate I. Bromination of I by N-bromosuccinimide (NBS) gave the dibromide monomer II. Polymers 1a and 1b were synthesized by Stille coupling reaction of 4,7-dibromo-2,1,3-benzothiadiazole with 3,5-di-n-decanyldithieno[3,2-b:2′,3′-d]thiophene 2,6-ditin and N-n-dodecyldithieno[3,2-b:2′,3′-d]pyrrole 2,6-ditin, respectively. Polymers 2a and 2b were synthesized by Stille coupling reaction of II with 3,5-di-n-decanyldithieno[3,2-b:2′,3′-d]thiophene 2,6-ditin and N-n-dodecyldithieno[3,2-b:2′,3′-d]pyrrole 2,6-ditin, respectively.

Scheme 1.

Synthesis of the polymers.

All these polymers have good solubility in common organic solvents such as chloroform, THF, and toluene. All the polymers can readily be processed to form smooth and pinhole-free films upon spin-coating except 1a. Molecular weights of the polymers were determined by gel permeation chromatography (GPC) using polystyrene standards as calibrants with toluene as eluent. Polymers 1a-b and 2a-b show Mn values of 5500, 44,800, 85,500, and 9600, respectively, (Table 1). The relatively low molecular weight of 1a may be due to the steric hindrance from 3,5-dialkyl substituents on the DTT unit. Thermogravimetric analysis (TGA) was used to characterize the thermal properties of the polymers under nitrogen. Polymers 1b and 2a-b have good thermal stability with decomposition temperatures (Td) of 326, 347, and 295 °C, respectively, whereas polymer 1a exhibits poor thermal stability with a Td of 160 °C (Fig. 2, Table 1). The relatively low molecular weight of 1a should be responsible for its poor thermal stability and film-forming property.

Figure 2.

TGA curves of the polymers.

Table 1. Molecular Weights and Thermal Properties of the Polymers
PolymerYield (%)MnaMwaPDIaTdb (°C)
  • a

    Number-average molecular weight (Mn), weight-average molecular weight (Mw) and polydispersity index (PDI) determined by means of GPC with toluene as eluent on the basis of polystyrene calibration.

  • b

    Onset decomposition temperature estimated using TGA under N2.

1a455,5006,7001.2161
1b7844,80082,3001.8326
2a8685,500123,4001.4347
2b84960012,8001.3295

Optical Properties

The UV-vis spectra of the copolymers in dilute (10−6 M) chloroform solution and thin films are shown in Figure 3. The absorption maxima of the D-A copolymers as well as other reported fused thiophene-benzothiadiazole copolymers are summarized in Table 2 for comparison. In solution, all polymers exhibit two absorption bands; the low energy band is due to ICT between the donor and the acceptor. Polymers 1a-b exhibit low energy absorption at 490 and 684 nm, respectively. The significant blue shift of absorption of 1a suggests highly twisted structure and limited conjugation of the polymer chain caused by the steric hindrance of 3,5-didecanyl-DTT moiety. Similar phenomena were observed in head-to-head polyalkylthiophenes,38 poly(3,6-dinonylthieno-[3,2-b] thiophene),39 and poly(3,5-didecanyl-DTT).23 The low energy absorption maximum of 1b is more or less close to those reported for similar polymers based on DTC (1d)5 and DTS (1f),13 but surprisingly, significantly blue shifts (80 nm) compared with the very similar DTP polymer (1c)29 (Table 2).

Figure 3.

UV-vis spectra of the polymers in chloroform (left) and in film (right).

Table 2. Low-Energy Absorption Maxima (nm) of the Polymers in Solution and in Film
PolymerFused RingaSolutionbFilmRef.
  • a

    Fused rings in Figure 1.

  • b

    In chloroform unless otherwise noted.

  • c

    In THF.

1aDTT490494This work
1bDTP684706This work
1cDTP76477129
1dDTC718c7655
1fDTSca. 670ca. 69013
2aDTT594626This work
2bDTP626722This work
2cDTP67169728
2dDTC61569637
2eDTS563c59336

The absorption spectrum of 2a in solution peaked at 434 and 594 nm is much broader and red shifts ∼100 nm compared with that of 1a (330 and 490 nm). This significant red shift mainly benefits from reduced steric hindrance of the backbone of 2a. The significant red-shift in the low-energy band seen in 2b on replacing DTT in 2a with DTP is consistent with a charge-transfer assignment given the more electron-rich character of DTP versus DTT.40 However, compared with that of 1b, the low-energy band of 2b blue shifts 58 nm, due to the weakened ICT effect of the D-A-D bridged by 3-dodecylthiophene unit. Interestingly, although 2b and 2c28 have very similar main chains, the difference of their low-energy bands is as large as 45 nm (Table 2). Again, the blue-shifted absorption of 2b is ascribed to its somewhat twisted main chain caused by the steric hindrance of 3-dodecylthiophene moiety. In comparison with 2d37 and 2e36, the low-energy band in 2b red shifts, consistent with the more electron-rich character of DTP versus DTC and DTS (Table 2).

The absorption of polymer 1a in thin film is almost identical to that in solution, which is related to the highly twisted polymer chain (Fig. 3). In contrast to polymer 1a, polymers 1b, and 2a-b exhibit significantly red-shifted (22–96 nm) and broadened absorptions in the solid state with respect to those in solution, indicating strong intermolecular interaction in solid state. Similar changes in UV-vis spectra have been reported for similar polymers 1c,51e,132c-e28, 36, 37 (Table 2). Polymers 1b and 2a-b show significant absorptions throughout the visible and extending into the NIR region (from 300 to 1000 nm). In particular, the absorption at 600–900 nm is strong, indicating good overlap with the photon flux of the sun peaked at 700 nm. The optical bandgap deduced from the absorption edge in film is 1.2, 1.5, and 1.3 eV for 1b, 2a-b, respectively, much smaller than that (1.9 eV) of widely used regioregular poly(3-hexylthiophene) (P3HT).

Among these four copolymers, only 1a emits strong photoluminescence (PL). The normalized PL spectra of 1a in solution and film are shown in Figure 4. 1a shows red emission with a same peak at 648 nm both in dilute solution and thin film. Both the absorption and fluorescence spectra of polymer 1a keep almost unchanged in the solution and in thin film, suggesting that the π–π stacking among the polymer chains is limited due to the highly twisted structure of the main chain.

Figure 4.

Fluorescence spectra of polymer 1a in solution and in film.

Electrochemistry

To estimate HOMO and LUMO energy levels of the polymers, we studied electrochemical properties using cyclic voltammetry of films drop cast onto glassy carbon working electrodes; an example of a cyclic voltammogram is illustrated in Figure 5. 2a shows one quasi-reversible oxidation peak and one quasi-reversible reduction peak. Other three polymers exhibit one irreversible oxidation peak and one quasi-reversible reduction peak. The estimated LUMOs for 1a-b and 2a-b (−2.9 to −3.3 eV, Table 3) as well as the HOMOs for 1b and 2a-b (−4.6 to −5.1 eV, Table 3) are similar to one another, and also are reminiscent of those of similar polymers with DTP,28, 29 DTC,11 and DTS13, 36 groups in the main chains. However, the HOMO for 1a is significantly lower than those of other polymers, due to highly twisted structure and limited conjugation of the polymer chain. As in several similar polymers with alternating DTP and benzothiadiazole28 or benzo[1,2-c;4,5-c′]bis[1,2,5]thiadiazole units,2 while the HOMO wave function is delocalized over the whole π system and the LUMO wave function is localized on the acceptor units.

Figure 5.

Cyclic voltammogram of 2a in CH3CN/0.1 M [nBu4N]+[PF6] with ferrocenium/ferrocene as an internal standard, at 50 mV s−1. The horizontal scale refers to an anodized Ag wire pseudoreference electrode.

Table 3. Redox Potentials and Energy Levels of the Polymersa
PolymerFused RingbEoxc (eV)Eredc (eV)HOMO (eV)LUMO (eV)Ref.
  • a

    Thin films in CH3CN/0.1 M [nBu4N]+[PF6], versus ferrocenium/ferrocene at 50 mV s−1.

  • b

    Fused rings in Figure 1.

  • c

    Eox is the onset potentials corresponding to oxidations, whereas Ered is the onset potentials corresponding to reductions.

  • d

    HOMO and LUMO estimated from the onset oxidation and reduction potentials, respectively, assuming the absolute energy level of ferrocene/ferrocenium to be 4.8 eV below vacuum.

1aDTT0.80−1.52−5.60d−3.28dThis work
1bDTP−0.15−1.67−4.65d−3.13dThis work
1cDTP  −4.81−3.0829
1eDTC  −5.3−3.5711
1fDTS  −5.05−3.2713
2aDTT0.32−1.90−5.12d−2.90dThis work
2bDTP0.27−1.67−5.07d−3.13dThis work
2cDTP  −5.0−3.4328
2eDTS  −5.13−3.2336

Field-Effect Transistors

OFET devices were fabricated in a top-contact configuration using Au as source and drain electrodes to study charge transport properties of these polymers. For polymer 1a, the quality of the films spin coated from toluene, chlorobenzene, or o-dichlorobenzene is too bad to measure its OFET properties. Polymer 1b also shows no signal in OFET device. The OFET device based on 2a exhibits a hole mobility of 8 × 10−5 cm2 V−1 s−1 after thermal annealing at 100 °C. Polymer 2b exhibits a mobility of 4 × 10−3 cm2 V−1 s−1 with current on/off ratio of 3 × 102 at room temperature; thermal annealing at 100 °C leads to improved performance with a mobility as high as 3 × 10−2 cm2 V−1 s−1 and current on/off ratio of 103 (Fig. 6). The mobility of 2b is more than two orders of magnitude higher than that of 2a, possibly due to more ordered film structure induced by the long substituent on N atom. The mobility of 3 × 10−2 cm2 V−1 s−1 for 2b is higher than those of DTC-based counterpart (2d)37 and DTS-benzothiadiazole copolymer (1f),13 and is similar to that of DTC-benzothiadiazole copolymer (1e)11 (Table 4).

Figure 6.

Current-voltage characteristics (IDS versus VDS) at different gate voltages (VGS) (left) and −IDS and (–IDS)1/2 versus VGS plots at VDS of –100 V (right) for a top contact device (annealing at 100 °C, W = 3 mm, L = 50 μm, 100 nm of 2b).

Table 4. The Performance of OFETs Based on the Polymers
PolymerFused Ringaμh (cm2 V−1 s−1)On/Off RatioVthb (V)Ref.
  • a

    Fused rings in Figure 1.

  • b

    Threshold voltage.

2aDTT8 × 10−56 × 103−10This work
2bDTP3 × 10−21 × 103−16This work
2dDTC2 × 10−437
1dDTC0.171055
1eDTC2 × 10−211
1fDTS3 × 10−313

Atomic force microscopy (AFM) measurements were used to study the relationship between the morphology of the films and their OFET properties (Supporting Information, Figure S1). The AFM image of 2a indicates that this polymer film is very rough and not continuous and homogeneous (roughness of 25.2 nm), which is responsible for the low mobility. Compared with 2a, the film of polymer 1b is relatively smooth and homogeneous with a roughness of 13.9 nm; no any crystalline or self-organization domains were observed, indicating an amorphous morphology. The OFET devices based on 1b did not show OFET signal for unknown reason. For polymer 2b before annealing, the film is more homogeneous and smooth (roughness of 10.4 nm) than that of 1b. It is interesting that the film exhibits domains of self-organization, which in turn enhances ordered structure formation and charge transport in the thin film.41 When annealed at 100 °C, the film roughness became smaller (7.8 nm), which led to higher mobility of the device.

Photovoltaic Cells

To demonstrate potential applications of these polymers in PSCs, we chose polymers 2a-b as examples to fabricate PSC devices because they possess relatively high mobility and broad absorption. We used 2a-b as an electron donor and solution-processable fullerene derivative PC60BM as an electron acceptor, and fabricated bulk heterojunction PSCs with a structure of ITO/PEDOT: PSS/2a-b:PC60BM (1:4, w/w)/Al. The current density (J) versus voltage (V) curves of the PSCs before and after annealing are shown in Figure S2 (Supporting Information); the open circuit voltage (Voc), short circuit current density (Jsc), fill factor (FF), and the PCE of the devices are summarized in Table 5. A PCE of 0.17 and 0.31% was obtained without annealing for 2a and 2b, respectively, (Table 5). The PCE of the PSC based on 2b:PC60BM is almost double that based on 2a:PC60BM, consistent with that 2b exhibits higher mobility. Thermal annealing at 150 °C led to a lower PCE (0.15 and 0.18% for 2a and 2b, respectively). Using PC70BM instead of PC60BM without annealing, the PCE decreased (0.21%). However, annealing at 110 °C led to improved device performance; Jsc, FF, and PCE increased from 1.74 mA/cm2, 0.3 and 0.21% to 2.88 mA/cm2, 0.37 and 0.43%, respectively, (Fig. 7).42 From AFM images of 2b:PC70BM (1:1, w/w) (Figure S3, Supporting Information), we observed that after annealing the blend film was much smoother (roughness decreased from 0.95 to 0.68 nm). Accordingly, thermal annealing enhanced the external quantum efficiency (EQE) of the PSC; the maximum EQE at 375–380 nm changed from 13 to 20% (Fig. 7). The PCEs of 0.31–0.43% are higher than that for the devices based on the DTS counterpart (2e) (0.18%),36 but lower than those of the DTP (2c)28 and DTC (2d)37 counterparts (2.1%). Optimization of the device structures, such as donor/acceptor weight ratio, thickness, annealing temperature, can be expected to substantially increase the PCE of the PSCs.

Figure 7.

EQE curves (left) and current density–voltage characteristics (right) of a device with the structure ITO/PEDOT:PSS/2b:PC70BM (1:1, w/w)/Al before and after annealing at 110 °C under the illumination of an AM 1.5 solar simulator, 100 mW cm−2.

Table 5. Photovoltaic Performance of the PSCs
PolymerPolymer/PCBM (w/w)AnnealingVoc (V)Jsc (mA/cm2)FFPCE (%)
2a1:4 (C60)no0.351.50.320.17
2a1:4 (C60)150 °C0.331.510.30.15
2b1:4 (C60)no0.442.340.30.31
2b1:4 (C60)150 °C0.32.220.270.18
2b1:1 (C70)no0.41.740.30.21
2b1:1 (C70)110 °C0.42.880.370.43

CONCLUSIONS

Low bandgap conjugated copolymers of benzothiadiazole with DTT and DTP were synthesized by palladium(0)-catalyzed Stille coupling reaction. The fused rings and 3-alkylthiophene bridge exert significant impact on the thermal, optical, electrochemical, charge transport, and photovoltaic properties of these polymers. The highly twisted backbone of 3,5-dialkyldithienothiophene-benzothiadiazole copolymer 1a prevents the polymer chains from π–π stacking, and leads to relatively narrow absorption and strong emission, which are almost same in solution and film. In contrast to polymer 1a, polymers 1b, and 2a-2b with more coplanar backbones exhibit broad absorption spectra extending into the near-IR (from 300 to 1000 nm), the details of which can be varied either by incorporating 3-alkylthiophene bridge or by replacing DTT with the stronger donor, DTP. For polymers 1b and 2a-2b, the significantly red-shifted (22–96 nm) absorptions in the solid state with respect to those in solution, as well as no emission suggest strong intermolecular interaction in solid state. The highly twisted backbone of 1a leads to considerably higher oxidation potential, but affects the ease of reduction to a lesser extent, suggesting that the HOMOs are delocalized over the whole main chain, whereas the LUMOs are strongly benzothiadiazole-localized. OFET measurements on two of these materials show that mobility of 0.03 cm2 V−1 s−1 for the DTP-based polymer 2b is more than two orders of magnitude higher than that of its DTT counterpart 2a. These polymers can also be exploited in bulk-heterojunction PSCs in combination with fullerene-based acceptors; these devices show power conversion efficiencies up to 0.43% under simulated AM1.5.

EXPERIMENTAL

Materials

2,6-Bis(tri-n-butylstannanyl)-3,5-di-n-decanyldithieno[3,2-b:2′,3′-d]thiophene23 and 2,6-bis(tri-n- butylstannyl)dithieno[3,2-b:2′,3′-d]thiophene20 were synthesized according to our published procedures. N-n-dodecyl-2,6-bis(tri-n-butylstannyl) dithieno [3,2-b:2′,3′-d]pyrrole was synthesized according to literature method.26 Toluene was distilled from sodium-benzophenone under nitrogen before use. All other reagents were used as received.

Characterization

The 1H and 13C-NMR spectra were measured on a Bruker AVANCE 400 MHz spectrometer using tetramethylsilane (TMS; δ = 0 ppm) as an internal standard. Mass spectra were measured on a GCT-MS micromass spectrometer using the electron impact (EI) mode or on a Bruker Daltonics BIFLEX III MALDI-TOF Analyzer using MALDI mode. Elemental analyses were carried out using a FLASH EA1112 elemental analyzer. Solution (chloroform) and thin-film (on quartz substrate) UV-vis absorption spectra were recorded on a JASCO V-570 spectrophotometer. Emission spectra in chloroform and thin film were collected on a Hitachi F-4500 spectrofluorophotometer. Electrochemical measurements were carried out under nitrogen on a deoxygenated solution of tetra-n-butylammonium hexafluorophosphate (0.1 M) in acetonitrile using a computer-controlled Zahner IM6e electrochemical workstation, a glassy-carbon working electrode coated with polymer films, a platinum-wire auxiliary electrode, and an Ag wire anodized with AgCl as a pseudoreference electrode. Potentials were referenced to the ferrocenium/ferrocene (FeCpmath image) couple by using ferrocene as an internal standard. TGA measurements were performed on Shimadzu thermogravimetric analyzer (model DTG-60) under a nitrogen flow at a heating rate of 10 °C/min. The GPC measurements were performed on a Waters 515 chromatograph connected to a Waters 2414 refractive index detector, using toluene as eluent and polystyrene standards as calibrants. Three Waters Styragel columns (HT2, 3, 4) connected in series were used. AFM images were obtained on Nanoscope IIIa (Digital Instruments) in tapping mode. The thin films of polymers 1a-b and 2a-b were spin coated on Octadecyltrichlorosilane (OTS)-treated Si/SiO2 substrates. The 2b:PC70BM (1:1, w/w) blend film preparation conditions for the AFM measurements were kept the same as PSC device fabrication for accurate comparison.

Fabrication and Characterization of Field-Effect Transistors

FET devices were fabricated with a top-contact configuration. A heavily doped n-type Si wafer with a SiO2 layer of 500 nm and a capacitance of 7.5 nF cm−2 was used as the gate. OTS was used as a self-assembled surface modifier for SiO2. A 100 nm-thick (±10 nm) semiconductor film was spin-coated on top of the OTS-treated SiO2 from 10 mg/mL o-dichlorobenzene solution of the polymers. Gold source and drain contacts (50 nm) were deposited on the organic layer through a shadow mask under high vacuum. The channel length (L) and width (W) were 50 μm and 3 mm, respectively. All the measurement of electric property was carried out at ambient condition and room temperature using a Keithley 4200 SCS semiconductor parameter analyzer. Device annealing was carried out at 100 °C for 1 h in a vacuum oven under a pressure of 0.1 Pa.

Fabrication and Characterization of Polymer Solar Cells

The PSC structure used in this study was ITO/PEDOT:PSS/2a-b:PCBM/Al. The ITO-coated glass substrates were cleaned by ultrasonic treatment in deionized water, acetone, detergent, and isopropyl alcohol sequentially, followed by spin coating of a 30 nm layer of PEDOT:PSS (Baytron P from H. C. Starck). After drying PEDOT:PSS at 120 °C for 20 min, a ∼70 nm layer of 2a-b:PCBM (1:4, w/w) was spin coated from o-dichlorobenzene solution onto the top of PEDOT:PSS. The film thickness was verified by an Ambios Tech. XP-2 profilometer. Then the metal cathode Al (∼100 nm), was deposited on the active layer by vacuum evaporation under 3 × 10−5 Pa through a shadow mask. The active area of the device was 4 mm2. The current-voltage (I-V) measurement of the devices was conducted on a computer-controlled Keithley 236 Source Measure Unit. A xenon lamp with AM1.5 filter was used as the white light source, and the optical power at the sample was 100 mW cm−2. The light intensity at each wavelength was calibrated with a standard single crystal Si photovoltaic cell. To investigate the effect of thermal annealing on device performance, the devices were annealed at 150 °C for 15 min (2a-b:PC60BM) or at 110 °C for 10 min (2b:PC70BM). For EQE measurement, a xenon lamp (Oriel, model 66,150, 75 W) was used as light source, and a chopper and lock-in amplifier were used for phase sensitive detection. The wavelength was controlled using a monochromator. To ensure an accurate counting of incident photons, a calibrated Si photodiode was used as a reference device.

Synthesis

4,7-Bis(4-n-dodecylthiophen-2-yl)-2,1, 3-benzothiadiazole (I)

To a 100 mL three-neck round bottom flask, 5-tri-n-butylstannanyl-3-n-dodecylthiophene (15 mmol, 8.1 g), 4,7-dibromo-2,1,3-benzothiadiazole (5 mmol, 1.5 g), and dry toluene (50 mL) were added. The mixture was deoxygenated with nitrogen for 30 min. Pd(PPh3)4 (0.4 mmol, 0.46 g) was added under nitrogen. The mixture was refluxed overnight and then cooled down to room temperature. A solution of KF (15 g) in water (40 mL) was added and stirred at room temperature for 2.5 h to remove the tin impurity. The mixture was extracted with CHCl3 (2 × 200 mL), washed with water (2 × 250 mL), and dried over anhydrous MgSO4. After removing the solvent, the residue was purified by column chromatography on silica gel using petroleum ether as eluent yielding a red solid (2.7 g, 84%). 1H-NMR (400 MHz, CDCl3): δ 7.98 (d, J = 0.8 Hz, 2H), 7.82 (d, J = 3.0 Hz, 2H), 7.04 (s, 2H), 2.69 (t, J = 7.7 Hz, 4H), 1.70 (m, 4H), 1.26 (m, 36H), 0.88 (t, J = 6.8 Hz, 6H). 13C-NMR (100 MHz, CDCl3): δ 152.66, 144.37, 139.01, 129.02, 126.05, 125.52, 121.52, 31.92, 30.66, 30.51, 30.21, 29.68, 29.49, 29.36, 28.94, 28.66, 22.68, 14.10. HRMS (EI): m/z 636.3611 (calcd for C38H56N2S3, 636.3606). Anal. Calcd for C38H56 N2S3: C, 71.64; H, 8.86; N, 4.40. Found: C, 71.76; H, 8.91; N, 4.38%.

4,7-Bis(4-n-dodecyl-5-bromo-thiophen-2-yl)-2, 1,3-benzothiadiazole (II)

A mixture of I (1.1 mmol, 0.7 g), chloroform (15 mL), and AcOH (15 mL) in 100 mL round bottom flask wrapped with aluminum foil was cooled to 0 °C. NBS (2.8 mmol, 0.5 g) in DMF (4 mL) was added dropwise over a period of 5 min. The mixture was stirred at 0 °C for 0.5 h, then warmed up to room temperature and stirred overnight. The mixture was poured into 2 M NaOH aqueous solution (150 mL), extracted with chloroform (50 mL ×2), washed with water (200 mL ×2), and dried over anhydrous MgSO4. After removing the solvent, the residue was purified by column chromatography on silica gel using petroleum ether as eluent yielding a red solid (0.6 g, 72%). 1H-NMR (400 MHz, CDCl3): δ 7.73 (s, 2H), 7.66 (s, 2H), 2.62 (t, J = 7.7 Hz, 4H), 1.67 (m, 4H), 1.26 (m, 36H), 0.88 (t, J = 6.8 Hz, 6H). 13C-NMR (100 MHz, CDCl3): δ 152.14, 143.03, 138.47, 128.03, 125.20, 124.85, 124.70, 111.59, 31.92, 29.98, 29.93, 29.75, 29.67, 29.60, 29.44, 29.35, 29.29, 22.68, 14.10. MS (MALDI): m/z 794 (M+). Anal. Calcd for C38H54Br2N2S3: C, 57.42; H, 6.85; N, 3.52. Found: C, 57.47; H, 6.82; N, 3.59%.

Poly[(2,1,3-benzothiadiazole-4,7-diyl)-alt-(3,5-di- n-decanyldithieno[3,2-b:2′,3′-d]thiophene-2,6-diyl)] (1a)

To a 25 mL three-neck round bottom flask, 2,6-bis(tri-n-butylstannanyl)-3,5-di-n-decanyldithieno [3,2-b:2′,3′-d]thiophene (0.33 mmol, 348 mg), 4,7-dibromo-2,1,3-benzothiadiazole (0.33 mmol, 97 mg), and dry toluene (10 mL) were added. The mixture was deoxygenated with nitrogen for 30 min. Pd(PPh3)4 (0.04 mmol, 46 mg) was added under nitrogen. The mixture was refluxed for 3 days and then cooled down to room temperature. A solution of KF (5 g) in water (10 mL) was added and stirred at room temperature for 2.5 h to remove the tin impurity. The mixture was extracted with CHCl3 (2 × 200 mL), washed with water (2 × 250 mL), and dried over anhydrous MgSO4. After removing the solvent, the residue (2 mL) was dropped into methanol (150 mL). The precipitate was filtered out, Soxhlet extracted with methanol for 48 h and dried under vacuum to give a red solid (90 mg, 45%). 1H-NMR (400 MHz, CDCl3): δ (ppm) 7.80–7.68 (br, 2H), 3.00–2.76 (br, 4H), 1.86–1.76 (br, 4H), 1.43–1.22 (br, 28H), 0.87 (br, 6H). Mn, 5 500, PDI, 1.2.

Poly[(2,1,3-benzothiadiazole-4,7-diyl)-alt-(N-n-dodecyldithieno[3,2-b:2′,3′-d]pyrrole-2,6-diyl)] (1b)

To a 25 mL round-bottom flask, 4,7-dibromo-2,1,3-benzothiadiazole (0.34 mmol, 100 mg), N-n-dodecyl-2,6-bis(tri-n-butylstannyl)dithieno[3,2-b:2′, 3′-d]pyrrole (0.34 mmol, 315 mg), and dry toluene (10 mL) were added. The mixture was deoxygenated with nitrogen for 30 min. Then Pd(PPh3)4 (26 μmol, 30 mg) was added under nitrogen. The mixture was heated up to 110 °C and stirred for 3 days. After cooled to room temperature, the reaction mixture was added chloroform (100 mL) and washed with water. The organic layer was dried over anhydrous MgSO4, concentrated to 5 mL, and dropped into methanol (200 mL). The precipitate was filtered, and Soxhlet extracted with methanol for 24 h. Finally, the polymer was purified by size exclusion column chromatography over Bio-Rad Bio-Beads S-X1 eluting with chloroform to afford a black solid (127 mg, 78%). 1H-NMR (400 MHz, CDCl3): δ 7.72 (br, 2H), 7.48 (br, 2H), 4.21 (br, 2H), 2.01 (br, 2H), 1.60–1.02 (br, 18H), 0.88 (br, 3H). Mn, 44,800, PDI, 1.8.

Poly{[4,7-bis(4-n-dodecylthiophen-5-yl)-2,1,3- benzothiadiazole]-alt-(dithieno[3,2-b:2′,3′-d] thiophene-2,6-diyl)} (2a)

To a 25 mL round-bottom flask, 2,6-bis(tri-n-butylstannyl)dithieno[3,2-b:2′,3′-d]thiophene (0.45 mmol, 348 mg), II (0.45 mmol, 357 mg), and dry toluene (10 mL) were added. The mixture was deoxygenated with nitrogen for 30 min. Then Pd(PPh3)4 (26 μmol, 30 mg) was added under nitrogen. The mixture was heated up to 110 °C and stirred for 3 days. After reaction mixture cooled to room temperature, chloroform (100 mL) was added to the reaction mixture and washed with water. The organic layer was concentrated to 5 mL and dropped into methanol (200 mL). The precipitate was filtered, and Soxhlet extracted with methanol for 24 h. Finally, the polymer was purified by size exclusion column chromatography over Bio-Rad Bio-Beads S-X1 eluting with chloroform to afford a black solid (320 mg, 86%). 1H-NMR (400 MHz, CDCl3): δ 8.0–7.2 (br, 4H), 6.89 (br, 2H), 2.81 (br, 4H), 1.74 (br, 4H), 1.24 (br, 36H), 0.88 (br, 6H). Mn, 85 500, PDI, 1.4.

Poly{[4,7-bis(4-n-dodecylthiophen-5-yl)-2,1,3- benzothiadiazole]-alt-(N-n-dodecyl-dithieno[3,2- b:2′,3′-d]pyrrole-2,6-diyl)} (2b)

To a 25 mL round-bottom flask, II (0.32 mmol, 252 mg), N-n-dodecyl-2,6-bis(tri-n-butylstannyl)dithieno[3,2-b:2′,3′-d]pyrrole (0.32 mmol, 296 mg), and dry toluene (10 mL) were added. The mixture was deoxygenated with nitrogen for 30 min. Then Pd(PPh3)4 (26 μmol, 30 mg) was added under nitrogen. The mixture was heated up to 110 °C and stirred for 3 days. After the reaction mixture cooled to room temperature, chloroform (100 mL) was added to the reaction mixture and washed with water. The organic layer was concentrated to 5 mL and dropped into methanol (200 mL). The precipitate was filtered, and Soxhlet extracted with methanol for 24 h. Finally, the polymer was purified by size exclusion column chromatography over Bio-Rad Bio-Beads S-X1 eluting with chloroform to afford a black solid (266 mg, 84%). 1H-NMR (400 MHz, CDCl3): δ 7.9–7.3 (br, 4H), 7.14 (br, 2H), 4.21 (br, 2H), 2.92 (br, 4H), 1.90 (br, 2H), 1.68 (br, 4H), 1.45–1.10 (br, 54H), 0.88 (br, 9H). Mn, 9600, PDI, 1.3.

Acknowledgements

This work was supported by the NSFC (Grants 20774104, 20721061), the MOST (Grants 2006AA 03Z220 and 2006CB932100), the Chinese Academy of Sciences, and the Ministry of Education (SRF for ROCS).

Ancillary