Cyclopentadithiophene Units for Organic Photovoltaic
Cell Applications
KUANG-CHIEH LI,1YING-CHAN HSU,2JIANN-T’SUEN LIN,2CHANG-CHUNG YANG,3 KUNG-HWA WEI,1HONG-CHEU LIN1
1
Department of Materials Science and Engineering, National Chiao Tung University, Hsinchu, Taiwan, Republic of China 2
Institute of Chemistry, Academia Sinica, Taipei, Taiwan, Republic of China 3
Energy and Environmental Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan, Republic of China
Received 5 December 2008; accepted 17 January 2009 DOI: 10.1002/pola.23312
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: Five novel conjugated copolymers (P1–P5) containing coplanar cyclopen-tadithiophene (CPDT) units (incorporated with arylcyanovinyl and keto groups in dif-ferent molar ratios) were synthesized and developed for the applications of polymer solar cells (PSCs). Polymers P1–P5 covered broad absorption ranges from UV to near infrared (400–900 nm) with narrow optical band gaps of 1.38–1.70 eV, which are compatible with the maximum solar photon reflux. Partially reversible p- and n-doping processes of P1–P5 in electrochemical experiments were observed, and the proper molecular design for highest occupied molecular orbital (HOMO)/lowest unoc-cupied molecular orbital (LUMO) levels of P1–P5 induced the highest photovoltaic open-circuit voltage in the PSC devices, compared with those previously reported CPDT-based narrow-band-gap polymers. Powder X-ray diffraction (XRD) analyses suggested that these copolymers formed self-assembled p-p stacking and pseudobilay-ered structures. Under 100 mW/cm2of AM 1.5 white-light illumination, bulk hetero-junction PSC devices containing an active layer of electron donor polymers P1–P5 mixed with electron acceptor [6,6]-phenyl C61 butyric acid methyl ester (PCBM) in the weight ratio of 1:4 were investigated. The PSC device containing P1 gave the best preliminary result with an open-circuit voltage of 0.84 V, a short-circuit current of 2.36 mA/cm2, and a fill factor of 0.38, offering an overall power conversion effi-ciency (PCE) of 0.77% as well as a maximal quantum effieffi-ciency of 23% from the external quantum efficiency (EQE) measurements.VVC2009 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 47: 2073–2092, 2009
Keywords: conjugated polymers; copolymerization; cyclopentadithiophene; heteroatom-containing polymers; polymer solar cell
INTRODUCTION
The developments of new materials to be used in organic optoelectronic devices such as polymeric solar cells (PSCs) have become dramatically
Journal of Polymer Science: Part A: Polymer Chemistry, Vol. 47, 2073–2092 (2009) V
VC2009 Wiley Periodicals, Inc.
Correspondence to: H.-C. Lin (E-mail: [email protected]. tw)
attractive because they represent a green and renewable energy alternative to fossil energy and nuclear power. In particular, the so-called bulk heterojunction (BHJ) concept1 has been estab-lished in thin films of organic solar cell devices utilizing electron-donating conjugated polymers blended with electron-accepting species, such as fullerenes,2(a) dicyano-based polymers,2(b,c) or n-type nanoparticles.3 For these purposes, several novel polymeric materials have been extensively studied over the past decade. For example, the regio-regular poly(3-hexylthiophene) (P3HT)4 and poly[2-methoxy-5-(30,70 -dimethyloctyloxy)-p-phenylenevinylene] (MDMO-PPV)5 possessed a highest power conversion efficiency (PCE) approaching 5.0% in PSCs. However, several groups proposed new polymeric structures as sub-stitutes for these polymers, since the disadvan-tages on the PSC performance were somehow re-stricted by their relatively large band gaps,6 which only absorbed part of the visible light and limited the utility of the sunlight.
To further improve the absorption properties of the conjugated polymers, the intramolecular charge transfer (ICT) interactions between elec-tron-donor (D) and electron-acceptor (A) moieties have been extensively applied to the develop-ments of novel narrow-band-gap conjugated poly-mers with better PSC performance, especially in the band-gap region of 1.4–1.8 eV.7–16 Among them, the derivatives of polyfluorene,7 thiophene-based,8 and arylamine-based9 represent promis-ing features havpromis-ing PCE values. However, besides band gaps, several characteristics of conjugated polymers, including highest occupied molecular orbital (HOMO)/lowest unoccupied molecular or-bital (LUMO) levels and carrier mobilities, need to be simultaneously optimized to achieve higher photovoltaic performance.10
Recently, to obtain longer conjugation lengths, more planar molecular geometries, and more rigid structures in p-conjugated polymers,11(a) novel heteroaromatic fused-ring derivatives, including cyclopentadithiophene (CPDT) units, have been widely investigated in PSCs. Kraak et al. first reported the structural unit of CPDT in 1968,12(a) and the later prepared CPDT-based poly-mers12(b,c),13showed relatively high conductivities due to more extensive p-conjugation lengths as compared with polythiophene and polyfluorene derivatives. Because of the high planarities, long conjugation lengths, narrow band gaps, and strong intermolecular p-p interactions of the CPDT units, CPDT-based polymers possessing
good conductive properties were found to be a powerful approach to optimize the PSC perfor-mance. Recently, the derivatives of CPDT copoly-mers showed very promising PCE results (1.14– 5.5%)14 and high carrier mobilities (102–101 cm2/Vs),14(c),15 which demonstrated that the syn-thesized ICT polymers possess both prominent properties of narrow band gaps and high carrier mobilities.
Up to now, very few investigations of CPDT-based polymers have been reported for the appli-cations of PSC performance. Although the band gaps of the reported derivatives of CPDT homo-polymers and cohomo-polymers were relatively low,11–16 their HOMO energy levels were apparently not low enough to produce air-stable polymers with relatively high open circuit potential (Voc) values
in the ultimate PSC devices, where the highest Vocvalues of the CPDT-based polymers were still
not over 0.65 V.14,15 It is noticeable that a well-known design to tune the HOMO and LUMO lev-els of conjugated polymers would be the introduc-tion of electron-withdrawing units, such as nitro, carboxy, and cyano groups, to the conjugated sys-tems.17 In 1991, Ferraris and Lambert reported that CPDT-based polymers, bearing electron-withdrawing keto groups at the bridging carbons, showed relatively low band gap values around 1.20 eV.16(a)On the other hand, another important observation was found that the electron-with-drawing cyano groups could decrease the HOMO level and thus to stabilize the neutral state of the conjugated system.17(a)
On the basis of this concept, two different moi-eties of CPDT derivatives, that is, 2,6-diarylene-cyanovinylene-CPDT (M1) and CPDT-4-one (M2), were utilized as acceptor monomers to synthesize CPDT-based copolymers P1–P5. Besides, to increase the solubility without causing any addi-tional twisting of the repeating units in the result-ing copolymers, 4-carbon position of compound 2 could be favorably functionalized by diethylhexyl substitutions to produce CPDT unit (3) as the do-nor monomer. Therefore, our donor-acceptor approaches utilized in these CPDT-based copoly-mers (P1–P5) achieve the absorption spectra in the visible range of 400–850 nm (with tailing up to around 1000 nm) in solid films and finely tuned HOMO and LUMO levels with narrow electro-chemical band gaps of 1.30–1.66 eV. In addition, after thermal annealing, the molecular configura-tions of the p-conjugated CPDT-based copolymers could clearly ensure that highly organized p-p stackings could be easily formed in these
fused-heteroaromatic molecular frameworks, which were confirmed by the powder X-ray diffraction (XRD) analyses. Compared with those reported CPDT-based polymers, our copolymers in this report showed much improved Voc values with a
highest open-circuit voltage up to 0.84 V as well as suitable electronic energy levels and good proc-essabilities for PSC applications. So far, the pre-liminary PSC performance of these structurally related copolymers showed the best PCE effi-ciency up to 0.77% while blended with [6,6]-phenyl C61butyric acid methyl ester (PCBM), with
a short circuit current density (Isc) of 2.36 mA/cm2,
an open circuit voltage (Voc) of 0.84 V, and a fill
fac-tor (FF) of 0.38 under AM 1.5 (100 mW/cm2). Although the results for the PCE efficiencies of these un-optimized PSCs are not sufficiently high enough, this research affords a new concept to enhance the Voc properties via the electron
donor-acceptor (D-A) design to offset the low Voc
draw-backs, which are generally encountered in narrow-band-gap CPDT-based conjugated polymers.
EXPERIMENTAL
Materials
All chemicals and solvents were used as received. Compounds 1 (cyclopenta[2,1-b:3,4-b0 ]dithiophen-4-one)18 and 2 (4H-cyclopenta [2,1-b:3,4-b0 ]di-thiophene)13(c) were synthesized according to the literature procedures. The synthetic routes of monomers 1–2 and polymers P1–P5 are shown in Schemes 1 and 2, and the synthetic procedures of their intermediates were described. Chemicals and solvents were reagent grades and purchased from Aldrich, ACROS, TCI, and Lancaster Chemi-cal. Toluene, tetrahydrofuran, and diethyl ether were distilled to keep anhydrous before use. Measurements and Characterization
1
H NMR spectra were recorded on a Varian Unity 300 MHz spectrometer using CDCl3solvents.
Ele-mental analyses were performed on a HERAEUS CHN-OS RAPID elemental analyzer. Transition temperatures were determined by differential scanning calorimetry (DSC; Perkin-Elmer Pyris 7) with a heating and cooling rate of 10 C/min. Thermogravimetric analyses (TGA) were con-ducted with a TA instrument Q500 at a heating rate of 20C/min under nitrogen. Gel permeation chromatography (GPC) analyses were conducted on a Waters 1515 separation module using
poly-styrene as a standard and THF as an eluent. UV-visible absorption and photoluminescence (PL) spectra were recorded in dilute chloroform solu-tions (106 M) on a HP G1103A and Hitachi F-4500 spectrophotometer, respectively. Solid films of UV-vis and PL measurements were spin-coated on a quartz substrate from chlorobenzene solu-tions with a concentration of 10 mg/mL. Cyclic voltammetry (CV) measurements were performed using a BAS 100 electrochemical analyzer with a standard three-electrode electrochemical cell in a 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) solution (in acetonitrile) at room tem-perature with a scanning rate of 50 mV/s. In each case, a carbon working electrode coated with a thin layer of these copolymers, a platinum wire as the counter electrode, and a silver wire as the quasi-reference electrode were used. Ag/AgCl (3 M KCl) electrode was served as a reference elec-trode for all potentials quoted herein. During the CV measurements, the solutions were purged with nitrogen for 30 s, and the redox couple ferro-cene/ferrocenium ion (Fc/Fcþ) was used as an
external standard. The corresponding HOMO and LUMO levels in copolymer films of P1–P5 were calculated from Eox/onset and Ered/onset values of
the electrochemical experiments. The LUMO value of PCBM was in accordance with the litera-ture data.19(b) Each onset potential in the CV measurements was defined by the intersection of two tangents drawn at the rising current and background current.
X-Ray Diffraction Characterization
Synchrotron powder XRD measurements were performed at beamline BL17A of the National Synchrotron Radiation Research Center (NSRRC), Taiwan, where the wavelength of X-ray was 1.33361 A˚ . The XRD data were collected using Mar345 image plate detector mounted or-thogonal to the beam with sample-to-detector dis-tance of 250 mm, and the diffraction signals were accumulated for 3 min. The powder samples were
packed into a capillary tube and heated by a heat gun, whose temperature controller is program-mable by a PC with a PID feedback system. The scattering angle h was calibrated by a mixture of silver behenate and silicon.
Fabrication of Hole- and Electron-Only Devices The hole- and electron-only devices in this study containing copolymers P1–P5: PCBM (1:4) blend film sandwiched between transparent ITO anode and cathode. The ITO glasses were first ultrasoni-cally cleaned in detergent, de-ionized water, ace-tone, and isopropyl alcohol before the deposition. After routine solvent cleaning, the substrates were treated with UV ozone for 3 min. In the hole-only device, the modified ITO surface was obtained by spin-coating a layer of poly (ethylene dioxythiophene): polystyrenesulfonate (PEDOT:PSS) (50 nm). After baking at 100 C for 1 h, the substrates were then transferred into
a nitrogen-filled glove box. The active layer was spin coated (spin rate¼ 500 rpm; spin time ¼ 40 s) on top of PEDOT:PSS and then dried in covered glass Petri dishes. The film thicknesses of the active layer were measured to be 370, 320, 260, 420, and 290 nm for P1, P2, P3, P4, and P5, respectively. Subsequently, a 15 and 120 nm thick of MoO3and aluminum was thermally evaporated
under vacuum at a pressure below 2.5 105torr through a shadow mask. The active area of the device was 0.0314 cm2. In the electron-only de-vice, the PEDOT:PSS layer was replaced with Cs2CO3, which has been used as an efficient
elec-tron injection layer. The modified ITO surface was obtained by spin-coating a layer of Cs2CO3 (2
nm). The film thicknesses of the active layer were measured to be 340, 240, 280, 260, and 460 nm for P1, P2, P3, P4, and P5, respectively. Subse-quently, a 40 and 70 nm thick of Ca and alumi-num was thermally evaporated under vacuum at a pressure below 2.5 105 torr through a shadow mask. The active area of the device was 0.0314 cm2.
Device Fabrication and Characterization of Polymer Solar Cells
The photovoltaic cell (PVC) device structure used in this study was a sandwich configuration of ITO/PEDOT:PSS/active layer/LiF/Al, where the active layer was made of electron donor polymers P1–P5 mixed with electron acceptor [6,6]-phenyl C61 butyric acid methyl ester (PCBM) in the
weight ratio of 1:4 (w/w). The PVC devices were fabricated according to the procedures similar to those of EL devices. An ITO-coated glass sub-strate was precleaned and treated with oxygen plasma before use. A thin layer (50 nm) of PEDOT:PSS was spin-coated on an ITO substrate and heated at 130 C for 1 h. Subsequently, the preliminary active layer was prepared by spin coating from composite solutions of P1–P5/PCBM (1:4 w/w) in chlorobenzene (10 mg/mL) on the top of the PEDOT:PSS layer. The spin rate was about 800 rpm, and the thickness of the active layer was typically ranged between 100 and 160 nm, unless the detailed thickness is specified. The PVC de-vices were completed by deposition with 1 nm of LiF and 120 nm of Al. The film thicknesses were measured by a profilometer (Dektak3, Veeco/ Sloan Instruments). For PVC measurements, I–V curves were recorded under a solar simulator with AM 1.5 irradiation (at 100 mW/cm2). A 300 W xenon lamp (Oriel, #6258) with AM 1.5 filter
(Oriel, #81,080 kit) was used as the white light source, and the optical power at the sample was 100 mW/cm2detected by Oriel thermopile 71,964. The I–V characteristics were measured using a CHI 650B potentiostat/galvanostat. The external quantum efficiency (EQE) was measured using a CHI 650B coupled with Oriel Cornerstone 260 monochromator. All PVC devices were prepared and measured under ambient conditions.
Synthesis
4,4-Bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene (3)
Compound 2 (2.0 g, 11.2 mmol) was dissolved in DMSO (50 mL), and then 2-ethylhexyl bromide (4.3 g, 22.4 mmol) was added and followed by po-tassium iodide (50 mg). The mixture was purged with nitrogen and cooled in an ice bath, and ground KOH (2.0 g) was added in portions. The resulting green mixture was vigorously stirred overnight at room temperature. Water was added after reaction, and the reaction mixture was extracted with dichloromethane. Consequently, the organic layer was separated and dried with magnesium sulfate. Solvent was removed under vacuum, and the crude product was purified by chromatography using hexane as eluent. Subse-quently, the pure compound was obtained as col-orless oil. Yield: 3.60 g (80%).
1 H NMR (ppm, CDCl3): d 7.13 (m, 2H), 6.91 (m, 2H), 1.86 (m, 4H), 0.93 (m, 18H), 0.73 (t, J ¼ 6.4 Hz, 6H), 0.59 (m, J ¼ 7.2 Hz, 6H). 2,6-Dibromo-4,4-bis(2-ethylhexyl)-4H-cyclo-penta[2,1-b:3,4-b0]dithiophene (4)
Compound 3 (3.5 g, 8.7 mmol) and NBS (3.1 g, 17.4 mmol) were dissolved in 50 mL of DMF. The resulting solution was stirred to react at room temperature under nitrogen overnight. Water (50 mL) was then added and the organic phase was extracted with dichloromethane (100 mL) twice, washed with water, and dried with magnesium sulfate. After that, the solvent was removed under reduced pressure to obtain the product. The crude product was purified by column chromatography with hexane to obtain pale yellow oil (4.10 g). Yield: 83%.
1
H NMR (ppm, CDCl3): d 6.92 (s, 2H), 1.80 (m,
4H), 0.94 (m, 18H), 0.76 (m, 6H), 0.60 (m, J ¼ 7.2 Hz, 6H).
4,4-Bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene-2,6-dicarbaldehyde (5)
To a solution of compound 4 (4.4 g, 7.8 mmol) in THF (80 mL), n-BuLi (2.5 M solution in hexane, 7.0 mL, 17.9 mmol) was added at 78 C. After stirring for 1 h, a solution of N-formylmorpholine (1.9 g, 16.4 mmol) in THF (20 mL) was added. Af-ter an additional stirring for 1 h at 78 C, the mixture was allowed to warm up to room temper-ature. Next, the mixture was hydrolyzed by 1 N HCl, and the final solution was extracted with dichloromethane. The organic layer was dried over magnesium sulfate and the solvent was evaporated. Afterward, the crude product was purified by column chromatography (silica gel, EA/hexane 1:10) to yield a yellow solid. Yield: 83%. 1 H NMR (ppm, CDCl3): d 9.88 (s, 2H), 7.61 (s, 2H), 1.94 (m, 4H), 0.96–0.86 (m, 18H), 0.71 (m, 6H), 0.55 (m, J ¼ 7.2 Hz, 6H). 2,6-Bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2- yl)-4,4-Bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene (8)
A solution of compound 4 (6.5 g, 11.6 mmol) in 150 mL of dry THF was stirred in a two-necked flask and cooled at 78 C while n-butyllithium (2.5 M solution in hexane, 29.0 mmol) was added dropwise under nitrogen atmosphere. After reac-tion for 2 h at78C, compound 7 (6.0 mL, 29.0 mmol) was added carefully to the mixture solution at 78 C and then the mixture was allowed to warm up to react at room temperature overnight. The final solution was acidified with 100 mL of 10% HCl solution and stirred for 45 min at room temperature. The solution was extracted by dichloromethane and the organic layer was dried over magnesium sulfate. After removing the sol-vent by rotavapor, the crude product was purified by column chromatography (silica gel, CH2Cl2/
hexane 1:2) to afford compound 8 (4.93 g). Yield: 65%. 1 H NMR (ppm, CDCl3): d 7.43 (s, 2H), 1.84 (m, 4H), 1.35 (m, 24H), 0.95–0.56 (m, 30H).13C NMR (ppm, CDCl3): d 160.95 (2C), 144.06 (2C), 131.86 (2C), 126.34 (2C), 83.89 (4C), 52.64 (2C), 43.17, 35.11 (2C), 33.84 (2C), 28.30 (2C), 27.42 (2C), 24.74 (8C), 22.75 (2C), 14.06 (2C), 10.55 (2C). MS (FAB): m/z [Mþ] 655; calcd m/z [Mþ] 654.4. Anal. calcd for C37H60B2O4S2: C, 67.89; H, 9.24; S, 9.80.
Found: C, 67.92; H, 9.52; S, 10.29.
M1
A mixture of compound 5 (2.6 g, 5.6 mmol), com-pound 6 (i.e., 1-bromophenylacetonitrile, 5.5 g, 28 mmol), and methanol (300 mL) were mixed in a 500 mL two-neck round-bottom flask at room tem-perature. A catalytic amount of potassium tert-butoxide in methanol was added into this mix-ture. After reaction for 24 h, the product was fil-tered and dried. Chromatography on silica gel eluted with CH2Cl2/hexane 1:4 afforded M1 as a
red solid (4.1 g). Yield: 90%.
1 H NMR (ppm, CDCl3): d 7.62–7.48 (m, 12H), 1.95 (m, 4H), 0.97–0.90 (m, 16H), 0.75–0.59 (m, 14H). 13C NMR (ppm, CDCl3): d 160.42 (2C), 140.25 (2C), 134.38 (2C), 134.25 (2C), 133.02 (2C), 132.25 (4C), 126.93 (4C), 126.67 (2C), 122.87 (2C), 118.23 (2C), 105.34 (2C), 54.21 (2C), 43.07, 35.29 (2C), 34.09 (2C), 28.43 (2C), 27.30 (2C), 22.73 (2C), 14.01 (2C), 10.62 (2C). MS (FAB): m/z [Mþ] 815; calcd m/z [Mþ] 814.1. Anal. Calcd for C43H46Br2N2S2: C, 63.39; H, 5.69; N, 3.44; S, 7.87.
Found: C, 63.58; H, 5.39; N, 3.55; S, 8.22.
2,6-Dibromocyclopenta[2,1-b:3,4-b0 ]dithiophen-4-one (M2)
The synthesis of compound M2 was also followed by the similar procedure of compound 4. Com-pound 1 (2.0 g, 10.4 mmol) was dissolved in 50 mL of dimethylformamide under nitrogen in the dark, and NBS (3.8 g, 20.8 mmol) was added grad-ually. The crude product was purified by column chromatography with CH2Cl2/hexane (1:1) to get
a purple solid (3.1 g). Yield: 85%.
1
H NMR (ppm, CDCl3): d 6.98 (s, 2H). 13C
NMR (ppm, CDCl3): d 182.56, 150.07 (2C), 143.54
(2C), 124.41 (2C), 113.95 (2C). MS (EI): m/z [Mþ] 350; calcd m/z [Mþ] 349.8. Anal. Calcd for C9H2Br2OS2: C, 30.88; H, 0.58; S, 18.32. Found:
C, 31.10; H, 0.71; S, 18.42.
General Procedure for the Syntheses of Copolymers P1–P5
The synthetic routes of polymers are shown in Scheme 2.20 All of the polymerization procedures were carried out through the palladium(0)-cata-lyzed Suzuki coupling reactions. In a 50 mL two-neck flask, 1 equiv of dibromo compounds (M1 and M2 with various molar ratios, M1:M2¼ m:0, 2:1, 1:1, 1:2, and 0:n, respectively) and 1 equiv of 2,6-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-
yl)-4,4-Bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene (8) were added into 10 mL of an-hydrous toluene. The Pd(0) complex, tetrakis(tri-phenylphosphine)palladium (1 mol %), was transferred into the mixture in a dry environ-ment. Then, 2 M aqueous potassium carbonate and a phase transfer catalyst, that is, aliquat 336 (several drops), were subsequently transferred via dropping funnel the previous mixture under nitro-gen. The reaction mixture was stirred at 85C for 2 days and then both excess amounts of iodoben-zene and phenylboronic acid, the end-cappers, dis-solved in 1 mL of anhydrous toluene were added and stirred for 4 h, respectively. The reaction mix-ture was cooled to 50 C and added slowly into a vigorously stirred mixture of methanol/water (10:1). The polymers were collected by filtration and reprecipitation from methanol. The crude polymers were further purified by washing with acetone for 3 days in a Soxhlet apparatus to remove oligomers and catalyst residues. The chlo-roform fractions (350–400 mL) were reduced to 40–50 mL under reduced pressure and were pre-cipitated in acetone and finally air-dried overnight. P1
Following the general polymerization procedure, compound 8 (1.0 equiv) and M1 (1.0 equiv) were used in this polymerization to acquire a black powder. Yield: 50%. 1 H NMR (ppm, CDCl3): d 7.67 (br, m, 14H), 1.99 (br, m, 8H), 1.02–0.67 (br, m, 60H). ELEM. ANAL. Calcd: C, 77.37; H, 7.83; N, 2.65; S, 12.15. Anal. Found: C, 77.89; H, 7.35; N, 2.77; S, 12.17. P2
Following the general polymerization procedure, compound 8 (1.0 equiv), M1 (0.67 equiv), and M2 (0.33 equiv) were used in this polymerization to attain a black powder. Yield: 80%.
1
H NMR (ppm, CDCl3): d 7.67 (broad), 7.05 (s),
1.98 (broad), 1.02–0.68 (broad). ELEM. ANAL.
Found: C, 74.97; H, 7.06; N, 2.65; S, 13.87; O, 0.96.
P3
Following the general polymerization procedure, compound 8 (1.0 equiv), M1 (0.5 equiv), and M2 (0.5 equiv) were used in this polymerization to obtain a black powder. Yield: 67%.
1
H NMR (ppm, CDCl3): d 7.65 (broad), 7.04 (s),
1.96 (broad), 1.00–0.66 (broad). ELEM. ANAL.
Found: C, 73.43; H, 6.97; N, 2.33; S, 15.01; O, 1.53.
P4
Following the general polymerization procedure, compound 8 (1.0 equiv), M1 (0.33 equiv), and M2 (0.67 equiv) were used in this polymerization to gain a black powder. Yield: 81%.
1
H NMR (ppm, CDCl3): d 7.66 (broad), 7.02 (s),
1.96 (broad), 1.25–0.69 (broad). ELEM. ANAL.
Found: C, 72.13; H, 7.59; N, 1.67; S, 16.30; O, 1.79.
P5
Following the general polymerization procedure, compound 8 (1.0 equiv) and M2 (1.0 equiv) were used in this polymerization to get a black powder. Yield: 56%. 1 H NMR (ppm, CDCl3): d 7.01 (br, s, 4H), 1.93 (br, m, 4H), 1.25–0.67 (br, m,30H). ELEM. ANAL. Calcd: C, 69.11; H, 6.48; S, 21.70; O, 2.71. Anal. Found: C, 69.73; H, 6.09; S, 21.04; O, 3.18.
RESULTS AND DISCUSSION
Syntheses and Chemical Characterization
As outlined in Scheme 1, two electron-accepting monomers M1 and M2 based on CPDT moieties were prepared from cyclopenta[2,1-b:3,4-b0 ]di-thiophen-4-one (1)18 using a reduction procedure and followed by dibromination, which were described by Turner and coworkers.13(c)The elec-tron-donating unit of CPDT (8) was prepared by dilithiation of 4 with n-buthyllithium and fol-lowed by reaction with compound 7 to afford 4,4- dialkyl-2,6-bis(4,4,5,5-tetramethyl-1,3,2-dioxabor-olan-2-yl)-4H-cyclopenta[2,1-b:3,4-b0]dithiophene 8 (see Scheme 1). Monomers M1–M2 and com-pound 8 were satisfactorily characterized by 1H NMR, 13C NMR, MS spectroscopies, and elemen-tal analyses. Three-component random copoly-mers P2–P4 were prepared successfully via Suzuki coupling of compound 8 with a mixture of various molar ratios of monomers M1 and M2. Two-component copolymers P1 and P5 were pro-duced by compound 8 copolymerized with mono-mers M1 and M2, respectively. The synthetic pro-cedures toward copolymers P1–P5 are outlined in Scheme 2. Most copolymers are partly soluble in organic solvents such as chloroform, THF, and chlorobenzene at room temperature and
completely soluble in high boiling point solvents (e.g., chlorobenzene) at high temperature. The yields and molecular weights of polymers P1–P5 determined by GPC against polystyrene stand-ards in THF are summarized in Table 1. These results show that considerable molecular weights with high yields (50–81% after Soxhlet extrac-tions) were obtained in these copolymers, where the weight-average molecular weights (Mw)
rang-ing 9700–60,800 with polydispersity indices (PDI ¼ Mw/Mn) of 1.41–2.65 were obtained.
The molecular structures of copolymers P1–P5 were identified by1H NMR and FT-IR. The output ratios of copolymers P2–P4 were calculated from the elemental analyses, which are all reasonably close to the feeding ratios of copolymers P2–P4. Proton NMR spectra of monomers M1–M2 and copolymers P1–P5 in CDCl3 are illustrated in
Figure 1. The characteristic resonances at 7.67 and 7.01 ppm in the spectra of P1–P5 are assigned to two different protons of monomers M1 and M2, respectively. In addition, the peak area ratios of output copolymers between the two reso-nances at 7.67 ppm (M1) and 7.01 ppm (M2) in the NMR spectra fitted well with the designed molecular structures of copolymers P1–P5, where a larger integrated signal of d ¼ 7.01 ppm could be observed in the copolymers with a higher molar ratio of M2. The molecular structures of polymers P1–P5 could also be confirmed by the FT-IR spectra. For instance, the absorption stretching mode of the cyano group in the copoly-mers, which typically appears at 2210/cm, was absent in the spectrum of copolymer P5, and the intensity of this band decreased as the molar ratio of M1 unit reduced from P1 to P4. In contrast, a characteristic band at 1710/cm for C¼¼O stretch-ing was observed in these copolymers, which was
absent in the spectrum of copolymer P1, and the intensity of this band increased as the molar ratio of M2 unit increased from P2 to P5.
The thermal stabilities and phase transition properties of copolymers P1–P5 were character-ized by TGA and DSC measurements under nitro-gen atmosphere, and the thermal decomposition temperatures (Td) and melting points (Tm) are
summarized in Table 1. It is apparent that all copolymers exhibited good thermal stabilities, which showed less than 5% weight loss upon heat-ing to 311–388 C. Regarding DSC experiments, there were no distinct glass transition tempera-tures (Tg) for all copolymers. Except for P1, these
copolymers showed relatively sharp transitions appearing around 192–229C, which were attrib-uted to the melting of the polymer backbones. The absence of sharp transition in P1 was probably originated from four 2-ethylhexyl irregular side chains belonging to monomers 8 and M1.
Optical Properties
The optical absorption spectra of D-A copolymers
P1–P5 in chloroform solutions (106M) and solid films are shown in Figure 2, and their photophysi-cal properties are demonstrated in Table 2. As can be seen, the absorption energy band gaps of CPDT-based copolymers P1–P5 could be finely tuned by the molar ratios of electron-accepting units M1 and M2 (M1:M2 ¼ m:0, 2:1, 1:1, 1:2, and 0:n), and their absorption spectra covered broad wavelength ranges for both solutions and solid films. The longer maximum absorption wavelengths of P1 (584 nm) and P5 (705 nm) in chloroform solutions of Figure 2(a) were about 88 nm and 192 nm red-shifted from the correspond-ing absorption wavelength of monomers M1 (496
Table 1. Molecular Weights, Yields, and Thermal Data of Polymers 1–5
Polymer Feeding Ratio (m:n) Output Ratio (m:n)a Mnb Mwb PDI Yield (%) Tmc(C) Tdd(C)
P1 m:0 m:0 15000 28800 1.92 50 n.de 388
P2 2:1 1.58:1 22900 60800 2.65 80 229 360
P3 1:1 0.87:1 14500 26300 1.81 67 192 355
P4 1:2 1:1.87 10200 17400 1.71 81 200 320
P5 0:n 0:n 6900 9700 1.41 56 200 311
aOutput molar ratios of m:n in copolymers P2–P4 were calculated from the elemental analyses. b
Molecular weights (Mn: number average molecular weight; Mw: weight average molecular weight) and PDI values were measured by GPC, using THF as an eluent, polystyrene as a standard.
cMelting transition temperatures (C) were measured by DSC at a heating rate of 10C/min.
dDecomposition temperatures (C) at 5% weight loss (T
d) were measured by TGA at a heating rate of 20C/min under nitro-gen.
eNo noticeable T
nm) and M2 (513 nm), respectively, reflecting much longer effective conjugation lengths of the extended coplanar CPDT-based polymer back-bones. However, it is noted that P1 exhibited one maximum absorption wavelength kmaxat 584 nm,
which was significantly longer (and had a longer conjugation length) than that of the related homo-polymer
poly(4,4-dialkyl-4H-cyclopenta[2,1-b:3,4-b0]dithiophen) (PCPDT with kmax ¼ 565 nm and
alkyl¼ 2-ethylhexyl).13(b)Similar trends of UV-vis spectra were observed in rigid conjugated poly-mers with strong ICT interactions between elec-tron donor and acceptor moieties.21 Surprisingly, the UV-vis spectrum of P5 displayed two well-sep-arated peaks at 484 nm and 705 nm, which were originated from two individual UV-vis absorption
peaks of 3 and M2 at 312 nm11(a) and 513 nm, respectively, before copolymerization. The shorter wavelength absorption in the region of 350–550 nm (484 nm) resulted from the incorporated do-nor unit (3) in copolymer P5, which was hypso-chromically shifted compared with the corre-sponding band of homopolymer PCPDT (kmax ¼
565 nm). Besides, the longer wavelength absorp-tion shoulder between 600 and 800 nm (ca. 705 nm) with tailing around 900 nm could be attrib-uted to the acceptor unit (M2) incorporated with the main chain of copolymer P5, which agreed well with those observed in the CPDT polymer derivatives containing the acceptor unit (M2).16 The main attribution of this effect can be explained by that the introduction of electron-defi-cient carbonyl moieties into the CPDT-based main-chain could also reduce the effective conju-gation length of the polymer backbone, and thus, to induce a hypsochromic shift of the absorption spectrum. This phenomena is also suggestive by the meta conjugation effect observed from amino-stilbenes11(b) and similar results with fluorene-CPDT-based copolymers.11(c) In other words, the electronic interaction between the carbonyl groups and the p-conjugated polymer backbones corresponds to the condition of meta-phenylene-bridged moieties.11(c) Therefore, copolymer P5 exhibits a more blue-shifted absorption maximum (484 nm) than that of homopolymer PCPDT (565 nm) because of the meta conjugation effect to prevent the p-electron delocalization by car-bonyl groups. Interestingly, reducing M1 contents and increasing M2 contents sequentially in copolymers P2, P3, and P4, gradual hypsochro-mic shifts of the short wavelength absorption
Figure 2. Normalized optical absorption spectra of D-A copolymers P1–P5 in (a) solutions (in chloro-form) and (b) solid films (spin-coating from chloroben-zene solutions).
Table 2. Photophysical Data in Chloroform Solutions and Solid Films and Optical Band Gaps of Polymers P1– P5
Polymer
kmax, UV (nm) kmax, PL (nm)
Dk (nm)a
Eg,opt(eV)b Solution Solid Filmc Solution Solid Film
P1 584 495, 620 653 724 36 1.70 P2 574 611 654 –d 37 1.59 P3 563 (704)e 609 654 –d 46 1.55 P4 544 (703) 582 (746) –d –d 38 1.73 (1.46) P5 484 (705) 520 (750) –d –d 36 1.95 (1.38) aDk
absorption¼ kmax, filmkmax, solution(nm).
bEstimated from the onset wavelength of UV-vis spectra of the thin solid film. cPL peaks were not detectable due to the PL quenching behavior.
dSpin-coated from chlorobenzene solution.
(560 nm) accompanying with slight increases of the longer shoulder absorption (700 nm) were observed in these copolymers. Hence, the intro-duction of electron-deficient carbonyl group in co-polymer P5 may reduce the effective conjugation length along the CPDT-based main chain because of the out of plane arrangements by the carbonyl groups of M2.
Figure 2(b) represents the UV-vis absorption spectra of solid films in the CPDT-based copoly-mers (P1–P5). The absorption spectra in solid films were generally similar to those in dilute so-lutions, where one maximum band in P1 was cen-tered at 620 nm and two characteristic bands in P2–P5 were centered at 520–611 nm (for the shorter wavelength absorption) and 746–750 nm (for the longer wavelength shoulder absorption), respectively. Because of the interchain association and p-p stacking of these copolymers in solids, the maxima of the p-p* transitions generally had lon-ger absorption maxima (36–46 nm of red shifts) in solid films than those in corresponding solutions. All copolymers (P2–P5) containing acceptor unit M2 had broad absorption bands that extended to the near-infrared region with a maximum absorp-tion shoulder kmax at750 nm, especially in P5.
The long tailing around 900 nm in the absorption spectra of P2–P5 could be observed in both solu-tions and solid films, which were attributed to their intrinsic properties rather than a reflection of poor film qualities. The optical band gaps (Eg,opt) of the copolymers in solid films, which
were determined by the cutoff absorption wave-lengths of the absorption spectra, are in the range of 1.38–1.70 eV (as shown in Table 2). As expected, the optical band gaps of all copolymers were not only much smaller than those of homo-polymer PCPDT13(b) and copolymers of poly(3-alkylthiophene)s,8(b),22 but also comparable with those of similar low band-gap copolymers, that is, poly(CPDT).12,14(c),21(a) Therefore, the idea of ICT interactions between electron donor and acceptor units in D-A copolymers is further supported by
an efficient method to narrow down the band gaps of the conjugated polymers,14(d),23which suggests that these copolymers can be useful materials for future photovoltaic applications.
The PL spectra of copolymers P1–P5 in chloro-form solutions and solid films excited at incident wavelengths of 500 nm and 550 nm, respectively, are shown in Figure 3. The PL emission spectra of the CPDT-based copolymers in solutions were dramatically quenched, which were enhanced by increasing the contents of M2 moieties in theD-A
copolymers (P1–P5) as shown in Figure 3(a). Interestingly, the PL spectra of copolymers P2– P5 containing M2 moieties in Figure 3(b) were completely quenched in solid films. The PL quenching phenomena of these polymers might stem from the intersystem crossing from the photo-excited singlet state to the triplet one was induced by the carbonyl group, where intramolec-ular (in solution) and intermolecintramolec-ular (in film) energy transfer along the conjugated main chain occurs. Additionally, the red shift of PL spectra of P1 from solution to film state might be due to the film morphology of highly crystallinity in P1 as supported by XRD analysis, which will be described in the XRD section later. The corre-sponding optical properties of these copolymers in solid films, including the broad and strong optical absorptions, propose their potential applications in PVCs described below.
Figure 3. Normalized PL spectra ofD-A copolymers
P1–P5 in (a) solutions (in chloroform) and (b) solid films (spin-coating from chlorobenzene solutions).
Electrochemical Characterization
The electronic states, that is, HOMO and LUMO levels, of the copolymers were investigated by CV to understand the charge injection processes in these new narrow-band-gap polymers and their PSC devices. The oxidation and reduction cyclic voltammograms of homopolymer PCPDT and copolymers P1–P5 in solid films are displayed in Figure 4. To obtain solid films of an equal thick-ness, the concentration in the THF solutions and film forming conditions were kept constant (5 mg/mL). The electrochemical measurements of the formal potentials, onset potentials, and band gaps, along with the estimated positions of the upper edges of the valence band (HOMO) and the lower edges of the conduction band (LUMO) are summarized in Table 3. As shown in Figure 4(a), the homopolymer PCPDT showed one reversible oxidation but no detectable reduction behavior, implying that the electrons are difficult to inject into this polymer. On the contrary, all copolymers P1–P5 exhibited one reversible oxidation and two reversible or quasi-reversible reduction peaks as evident from the areas and close proximity of the anodic and cathodic scans in Figure 4(b), which are a good sign for high structural stability in the charged state. As illustrated in Table 3, the formal oxidation potentials of these polymers were in the range of 0.74–1.05 V, and their formal reduction potentials were in the ranges of0.94 to 0.99 V and1.16 to 1.95 V, respectively.
The moderate onset oxidation potentials and onset reduction potentials of copolymers P1–P5 occurred between 0.5 and 0.85 V and about0.81 V, respectively, from which the estimated HOMO levels of 4.90 to 5.25 eV and LUMO levels of about3.59 eV were acquired according to the following equation:19(b),24EHOMO/LUMO¼ [(Eonset (vs. Ag/AgCl)Eonset (Fc/Fcþ vs. Ag/AgCl)4.8] eV, where
4.8 eV is the energy level of ferrocene below the vacuum level and Eonset (Fc/Fcþ vs. Ag/AgCl)¼ 0.4 eV.
In addition, the onset oxidation potential of homo-polymer PCPDT was observed at 0.55 V, from which the HOMO level of 4.95 eV was esti-mated. It is worthwhile to note that the HOMO energy levels of copolymers P1–P5 were signifi-cantly varied relative to that of homopolymer PCPDT as measured under the same condition. Compared with PCPDT, the HOMO energy levels of copolymers P4 to P1 were reduced gradually by 0.1–0.3 eV via the incorporation of the increasing amounts of electron-withdrawing cyano groups into the polymer backbones. There-fore, based on the oxidation potential data, the
higher contents of electron-withdrawing cyano groups in copolymers P1–P5 can induce the decreases in HOMO levels17(a)and show good air stabilities, especially for P1.25 However, the HOMO energy level of copolymer P5 was slightly higher than that of PCPDT (with a difference of 0.05 eV). It is probably that the electron-withdrawing effect of the ketone groups and the contribution of the primary resonance form might decrease the aromaticity of the system and hence to increase the quinoid character of the polymer backbones.16(a),26 In contrast, the electrochemical
reductions of copolymers P1–P5 showed similar LUMO energy levels at about3.59 to 3.60 eV, which represent to possess high electron affinities and also make these copolymers suitable donors for electron injection and transporting to PCBM acceptors (with 0.15–0.16 eV offsets in LUMO lev-els regarding PCBM with a LUMO level of3.75
Figure 4. Cyclic voltammograms of (a) homopoly-mer PCPDT and (b) copolyhomopoly-mers P1–P5 (thin solid films) at a scan rate of 100 mV/s.
eV,19as shown in Fig. 5) for the polymeric BHJ so-lar cell devices.27 Interestingly, the energy band gaps Eg,ec (Eg,ec ¼ Eox/onsetEred/onset, where Eg,ec
values are between 1.30 and 1.66 eV) measured directly from CV are close to the optical band gaps (Eg,opt between 1.38 and 1.70 eV) acquired from
the absorption spectra.
X-Ray Diffraction Analyses
To investigate the microstructural orders and mo-lecular arrangements of thermal annealed CPDT-based copolymers in solids, XRD measurements were performed on powder samples before and af-ter the thermal treatment at 150C. As shown in Figure 6, the annealed copolymers P1 and P5
both exhibited well-crystalline patterns, which indicate highly ordered arrangements in solids. Distinct primary diffraction peaks, including one peak at 2h ¼ 5.1 associated with a d-spacing value of 15.0 A˚ , were observed in copolymer P1 af-ter thermal annealing. Compared with P1, copoly-mer P5 exhibited substantially a primary
Table 3. Electrochemical Potentials, Energy Levels, and Band Gap Energies of Polymers P1–P5a
Polymer
Oxidation Potential Reduction Potential Energy Levelb Band Gap
V versus Ag/Agþ V versus Ag/Agþ eV eV
Eox/onset c Eox/o d Ered/onset c Ered/o d
EHOMO ELUMO Eg,ec
e PCPDT 0.55 0.74 N. Ae N. Ae 4.95 N. Ae N. Af P1 0.85 1.05 0.81 0.95 5.25 3.59 1.66 1.95 P2 0.74 1.04 0.81 0.97 5.14 3.59 1.55 1.28 P3 0.70 1.02 0.81 0.95 5.10 3.59 1.51 1.24 P4 0.65 0.84 0.81 0.94 5.05 3.59 1.46 1.24 P5 0.50 0.83 0.80 0.99 4.90 3.60 1.30 1.16
aReduction and oxidation potentials measured by cyclic voltammetry in solid films. b
Estimated from the onset potentials using empirical equations: EHOMO/ELUMO¼ [(Eonset (vs. Ag/AgCl)Eonset (Fc/Fcþ vs. Ag/ AgCl))4.8] eV where 4.8 eV is the energy level of ferrocene below the vacuum level and Eonset (Fc/Fcþ vs. Ag/AgCl)¼ 0.4 eV.
cOnset oxidation and reduction potentials. dFormal oxidation and reduction potentials. e
Eg,ec¼ Eox/onsetEred/onse.
fNo properties of cathodic reduction potentials were available.
Figure 5. Energy band diagram with HOMO/ LUMO levels of donor copolymers P1–P5 and PCBM acceptor in relation to the work functions of ITO and Al (HOMO value of PCBM was from literature19).
Figure 6. Powder XRD patterns of copolymers P1 (pristine and annealed samples) and P5 (annealed sample). The sharp diffraction peaks indicated that the polymers formed an order structure in the solid state.
diffraction feature with a wider angle at 2h ¼ 5.36 (corresponding to a smaller d-spacing value of 14.26 A˚ ), which was assigned to a distance between the conjugated backbones separated by the long side chains as reported for other similar p-conjugated polymers with long pendants.28The XRD diffraction patterns at 2h ¼ 10.2 and 10.7, related to the d-spacing values of 7.51 and 7.15 A˚ for copolymers P1 and P5, respectively, were the second-order peaks of the diffractions at 15.0 and 14.26 A˚ . Furthermore, copolymer P1 showed a higher crystalline characteristic with a diffraction peak up to the third-order at 2h ¼ 15.3, corre-lated to a d-spacing value of 5.0 A˚. Because the effective cross section (S) of polymer pendent alkyl chains is equal to 20 A˚ , the hexagonal-like aggregations of the alkyl chains showed a charac-teristic side-to-side distance between alkyl chains with d ¼ 4.2 A˚.29 The value observed for the dif-fraction feature at the d-spacing value of 4.22 A˚ in copolymer P1 is in agreement with the result as previously reported.29However, the hexagonal-like aggregation about d ¼ 4.2 A˚ was not observed in P5, which means that the alkyl side-chains in copolymer P5 have less crystalline behavior (only amorphous halo observed2h ¼ 18) and the hex-agonal-like aggregations of alkyl side-chains did not exist. Compared with copolymer P1, this lower packing order of the alkyl side-chains in P5 might be due to the lower packing density of alkyl pendants from M2 moieties in P5 than that from M1 moieties in P1. For the CPDT-based copoly-mers P1 and P5, the diffraction features at 2h ¼ 20.2 and 21.5, corresponding to the d-spacing values of 3.80 and 3.52 A˚ , respectively, are close to the layer-to-layer p-p stacking distances between the coplanar backbones of the reported p-conju-gated polymers15(b),28–30 and being somewhat larger than the sheet-to-sheet distance of graphite (3.35 A˚ ).29(a) The diffraction features of both copolymers P1 and P5 were often observed in the XRD patterns of the p-conjugated poly-mers.15(b),28–30 On the basis of the observation, it can be assumed that copolymers P1 and P5 form good p-p stackings consisting of p-conjugated co-planar backbones, but P1 has a better crystalline form in alkyl side chains than P5.
The possible packing motifs (side-view) of copolymers P1 and P5 are represented in Figure 7, which show a model that the alkyl side chains stack as bilayered packings and may have trivial interdigitated arrangements. It is interesting to note that the primary diffraction interchain dis-tance of copolymer P1 was somewhat (0.74 A˚)
larger than that of P5 from XRD data. As possible side-view packing motifs in Figure 7, the cyanovi-nylene and phecyanovi-nylene segments in the polymer backbones of copolymer P1 result in a more kinked molecular configuration with a wider p-p stacking region [5.59 E in Fig. 7(a)]. Compara-tively, due to the only simple CPDT-based moi-eties in copolymer P5, the comparatively linear backbones of copolymer P5 stack more compactly with a narrower rigid-core width [4.57 A˚ in Fig. 7(b)]. Since copolymers P1 and P5 have the same length of flexible tails, both copolymers might pre-fer the bilayered lamellar stacking in the soft regions with the same thickness of 5.2 A˚ 2 ¼ 10.4 A˚ . Therefore, the total lamellar thickness dif-ference of 0.74 A˚ in the diffraction interchain dis-tance of copolymers P1 and P5 from XRD data (15.0 and 14.26 A˚ for copolymers P1 and P5, respectively) was induced from the variation of their backbones’ widths in p-p stacking rigid-core regions, that is, 5.594.57 ¼ 1.02 A˚, where 5.59 A˚ and 4.57 A˚ are the rigid-core regions of copoly-mers P1 and P5, correspondingly. Moreover, the interchain lamellar d-spacing values of P1 and P5 (15.0 and 14.26 A˚ , respectively) from XRD are roughly equal to the total sum of the twice length of 2-ethylhexyl group plus the individual widths of their respective polymer backbones in the Chem3D ultra 8.0 calculations (15.99 and 14.97 A˚ , correspondingly) from the side-view of Figure 7. This result suggests that the side chains of the copolymers likely stack as bilayered structures in the lamellar sheets, though the precise orienta-tion of the alkyl side chains can not be determined with the present XRD information alone. The d-spacing values of 3.80 and 3.52 A˚ (obtained from XRD patterns at 2h ¼ 20.2 and 21.5) for copoly-mers P1 and P5 are correspondent to the (top-view) layer-to-layer p-p stacking distances between the top layer and bottom layer of the co-planar backbones in Figure 7(a,b), respectively. According to the XRD results, copolymer P1 has more and sharper XRD peaks to possess a better crystallinity than P5, especially for wide angles of (top-view) alkyl side-chain arrangements, where the hexagonal-like aggregation (ca. d ¼ 4.2 A˚) was only observed in P1. Overall, the proposed model can explain the possible structural arrangements of the copolymer chains in copolymers P1 and P5. Polymeric Photovoltaic Cell Properties
The motivation for the design and syntheses of the conjugated CPDT-based copolymers is to look
for new narrow-band-gap polymers for the appli-cation of PSCs. To investigate the potential use of copolymers P1–P5 in PSCs, BHJ devices were fabricated from an active layer in which
copoly-mers P1–P5 were blended with the complemen-tary fullerene-based electron acceptor, PCBM, in a weight ratio of 1:4 (w/w). PSC devices with a configuration of ITO/PEDOT:PSS/P1–P5:PCBM
Figure 7. Schematic representation of a proposed layered and p-p stacked copoly-mer structure in the Chem3D ultra 8.0 calculations of (a) P1 and (b) P5 in solid state. [Color figure can be viewed in the online issue, which is available at www. interscience.wiley.com.]
(1:4 w/w)/LiF/Al were fabricated by depositing a thin layer (50 nm) of PEDOT:PSS onto pat-terned ITO slides. The active layer (100–160 nm) consisting of P1–P5 and PCBM (1:4 w/w) was then deposited from a solution (10 mg/mL in chlorobenzene) by a spin rate of 800 rpm on the PEDOT:PSS film, and followed by the deposition of a LiF (1 nm) and aluminum (120 nm) back electrode. The PSC devices were measured under AM 1.5 illumination for a calibrated solar simula-tor with an intensity of 100 mW/cm2. The prelimi-narily obtained properties are summarized in Tables 4 and 5, and the typical I–V characteristics and EQE wavelength dependencies of all PSC devices are shown in Figure 8. Under the white-light illumination, the current density (Isc), open
circuit voltage (Voc), and FF of the PSC devices
composed of copolymers P1–P5 were in the range of 0.09–2.36 mA/cm2, 0.36–0.84 V, and 17–38%, respectively, with the PCE values between 0.01% and 0.77%.
The photovoltaic properties of the PSC devices containing CPDT-based copolymers P1–P5 were dependent on the solubility and film-forming quality of the copolymers. Among these PSC de-vices containing P1–P5, copolymer P1 gave the best performance in Figure 8(b) with Isc ¼ 2.36
mA/cm2, Voc ¼ 0.84 V, FF ¼ 38%, and PCE ¼
0.77%, respectively. Interestingly, the Isc value of
the PSC device containing P1 was strongly enhanced relative to those containing P2–P5 (by a factor of26 times higher than that of the worst P3), which might be due to the promoted solubil-ity and the better film-forming capabilsolubil-ity by add-ing a higher molar ratio of M1 units with alkyl side chains to P1. Ideally, the Isc values were
determined by the product of the photoinduced charge carrier densities and the charge carrier mobilities within the organic semiconductors.10(b) Thus, it can be recognized that the better results of Isc and FF in the PSC device containing P1
were obtained likely due to the well-balanced charge flow and less significant recombination loss4(c),9(b) originated from the highly ordered structural packing of alkyl side chains, as previ-ously proved by the XRD patterns in the wide angle region of P1. However, the relatively low Isc
and FF values in the PSC device containing P3 is poorly understood at this time, but it might be related to geminate charge recombination at the interface due to stable charge-transfer states, which limited the values of the photocurrents.31(a) Therefore, to further explore the dependence of charge transfer properties on the PSC devices, we
Table 4. Photovoltaic Properties of PSC Devices Containing an Active Layer of P1–P5:PCBM¼ 1:4 (w/w) with a Device Configuration of ITO/PEDOT:PSS/Polymer:PCBM/LiF/Ala
Active LayerbPolymer:PCBM Thickness (nm)b
Voc(V) Isc(mA/cm2) FF (%) PCE (%) P1 160 0.84 2.36 38 0.77 P2 140 0.48 0.77 23 0.08 P3 100 0.36 0.09 17 0.01 P4 140 0.49 0.67 25 0.08 P5 140 0.51 0.81 26 0.11
aMeasured under AM 1.5 irradiation, 100 mW/cm2. bP1–P5:PCBM¼ with the fixed weight ratio of 1:4 (w/w).
Table 5. Photovoltaic Propertiesaof Bulk-Heterojunction Solar Cells Containing an Active Layer of P1:PCBM¼ 1:4 (w/w) with Various Thicknesses
Thickness (nm)b
Spin Concentrations of Active Layer
(P1:PCBM) (mg/mL:mg/mL)c Voc(V) Isc(mA/cm2) FF (%) PCE (%)
120 5:20 0.77 0.42 15 0.05
160 10:40 0.84 2.36 38 0.77
310 20:80 0.83 1.46 25 0.31
aMeasured under AM 1.5 irradiation, 100 mW/cm2.
bThe thickness ( 10 nm) was controlled by the solution concentrations of the active layer P1/PCBM (1:4 by wt.), and the spin rate of the active layer (P1/PCBM) was fixed at ca. 800 rpm.
cThe active layer (P1/PCBM) was prepared from spin-coating of different solution concentrations (in chlorobenzene), but the weight ratio of P1:PCBM was fixed at 1:4.
have performed current measurements on hole-only and electron-hole-only devices. The electron and hole mobilities can be determined precisely by fit-ting the plot of the current versus the voltage (I– V) curves for single carrier devices to the SCLC model.31(b,c)These devices in this study containing copolymers P1–P5:PCBM (1:4) blend film sand-wiched between transparent ITO anode and cath-ode. The current is given by J ¼ 9e0erlV2=8L3,
where e0er is the permittivity of the polymer, l is
the carrier mobility, L is the device thickness. The best result of hole mobility was found to be 9.74 106 cm2/V/s for copolymer P1, and the others copolymers P2–P5 were found to be below 1.41 106cm2/V/s. Reasonably, copolymer P1 gave the best performance efficiency and highest photocur-rent property in the PSC devices. Additionally, the electron mobilities of copolymers P1–P5 were
found to be a range near4.78 105cm2/V/s. In comparison with the hole- and electron-mobilities of these copolymers in the blend system (poly-mers:PCBM ¼ 1:4), the electron-mobilities showed relatively fast charge transporting rates than that hole-mobilities because of larger PCBM amounts blended in the system. Therefore, it rev-els that the electron is the dominant charge car-rier in the PSC devices, which results in the unbalanced charge transport obtained in this study.
The Voc values were noticeably varied among
the PSC devices containing copolymers P1–P5, which were related to the differences between the HOMO energy levels of the polymers and the LUMO energy levels of the acceptors.10Therefore, the HOMO energy levels of the donor polymers in PSC devices are very important to be finely tuned for PSC devices with high efficiencies. As dis-cussed previously for the oxidation potentials of all copolymers, copolymer P1 incorporated with the electron-withdrawing cyano groups has the lowest HOMO level among copolymers P1–P5. Thus, the highest Vocvalue (0.84 V) is
satisfacto-rily reached in P1, which has the highest Voc
value for any reported CPDT-based materials so far. Surprisingly, followed by the decrease of the HOMO levels (see Fig. 5), the Voc values did not
comply with the previous general regulation in the results of PSC devices for P2–P5. However, the photovoltaic parameters could be also influ-enced to some extent by the thickness of the active layer.32 Especially for P3, although the co-polymer had a medium HOMO level, its PSC de-vice had the worst Vocvalue owing to a worse film
with a thinner thickness of about 100 nm induced by the poor solubility of copolymer P3.
To investigate the explanation for different effi-ciencies of the PSC devices, the EQE spectra of the PSC devices containing copolymers P1, P2, and P5 blended with PCBM (1:4 w/w) as the pho-tovoltaic layer are compared in Figure 8(b). The broad EQE curves of P1, P2, and P5 covered almost the entire visible spectrum from 350 to 700 nm with maximum EQE values of 23%, 5%, and 8% for P1, P2, and P5, respectively. In a detailed comparison, the PSC devices containing P1 and P2 exhibited photovoltaic responses at both 380 and 600 nm, but with a shoulder at 470 nm only for P1. However, the PSC device contain-ing P5 merely showed the maximum EQE values at 360 and 440 nm, but the longer wavelength shoulder absorption of 700–750 nm (as shown in Fig. 2) was not observed in the EQE spectra. The
Figure 8. (a) I–V curves of solar cells with active layers P1–P5:PCBM (1:4 w/w) under simulated AM 1.5 solar irradiation. (b) EQE wavelength dependen-cies of solar cell devices based on active layers P1:PCBM, P2:PCBM, and P5:PCBM (1:4 w/w). Inset: representative device configuration.
result shows that the unit of monomer M2 incor-porated into the polymer backbone can not partic-ipate in the generation of photocurrents and thus to result in a feature of absorption limitation, which can be explained by the Iscvalue of P5 was
relatively lower than that of P1. Comparing the PSC devices containing P2 and P5, the measured current and EQE properties in the region of P1 absorption comprised a wider wavelength range and a higher efficiency (with a maximal 4.6 times larger), which propose that P1 somehow contrib-uted significantly to the overall current generated by the (P1:PCBM)-based PSC device under illu-mination presumably owing to a more efficient intermolecular charge transfer.
Finally, the effect of varying the thickness of the active layer on the photovoltaic performance of P1-based PSC devices is explored as shown in Figure 9 and Table 5. The thicknesses of the active layers were varied in the range of 120–310 nm by changing the spin concentrations (5, 10, and 20 mg/mL) of P1 in chlorobenzene under the same spin rate. Quite surprisingly, decreasing the active layer thickness to 120 nm or increasing to 310 nm did not result in higher PCE efficiencies because there were simultaneous decreases in both FF and Iscvalues as revealed in Figure 9(a).
In contrast to the medium 160 nm thickness in the PSC device, both thicker (310 nm) and thin-ner (120 nm) devices showed slightly lower Voc
values but significantly reduced FF and Iscvalues,
where the thicker active layer had a combined influence on the hindered charge carrier transport or recombination33 and the thinner active layer reduces the absorption of the irradiated light. As shown in Figure 9(b), a similar tendency was also conceived in EQE spectra, where the PSC device with the medium thickness of 160 nm possessed a maximal EQE of 23% at the irradiation wave-length of 350–400 nm. The higher EQE values covering the broad absorption wavelength region further explain the improved PSC performance of the medium thickness device (160 nm) over the other two devices with thicker and thinner thick-nesses (310 nm and 120 nm). Additional improve-ments are underway to optimize the PSC devices by the modification of the film morphology, the process of thermal annealing treatments, and the replacement of some other electron acceptors, which can augment the formation of phase-sepa-rated structures and the charge mobilities.
CONCLUSIONS
Using the concept of incorporating electron-with-drawing groups in the D-A conjugated polymers, we have successfully synthesized five CPDT-based copolymers employing arylcyanovinyl and keto groups in different molar ratios by palladium(0)-catalyzed Suzuki coupling reactions. The band gaps and the HOMO/LUMO energy levels of these resulting copolymers can be finely tuned as dem-onstrated by the investigation of optical absorp-tion properties and electrochemical studies. In powder XRD measurements, these copolymers exhibited obvious diffraction features indicating a highly ordered p-p stacking in the solid state. Pre-liminary PSC devices based on these five
Figure 9. (a) I–V curves of solar cells under simu-lated AM 1.5 solar irradiation and (b) EQE spectra for PSC devices containing an active layer of P1:PCBM ¼ 1:4 (w/w) with three different thick-nesses (h) 120 nm, (l) 160 nm, and (*) 310 nm.
copolymers blended with PCBM acceptors (1:4 w/w) had the PCE up to 0.77%, which gave the best performance with the values of Isc¼ 2.36 mA/
cm2, FF ¼ 38%, and Voc ¼ 0.84 V. Furthermore,
this study provides novel conception that the HOMO energy levels can be reduced via the syn-theses of merging with electron-withdrawing func-tional groups and thus the open-circuit voltage can be considerably enhanced, which will significantly improve the low Voc values mainly possessed by
most CPDT-based narrow-band-gap polymers.
The authors express their sincere thanks to the National Center for High-performance Computing for computer time and facilities. The powder XRD meas-urements are supported by beamline BL17A (charged by Jey-Jau Lee) of the National Synchrotron Radiation Research Center (NSRRC), in Taiwan. They also acknowledge the financial supports of this project pro-vided by the National Science Council of Taiwan (ROC) through NSC 96-2113M-009-015, National Chiao Tung University through 97W807, and Chung-Shan Institute of Science and Technology (in Taiwan).
REFERENCES AND NOTES
1. Yu, G.; Gao, J.; Hummelen, J. C.; Wudl, F.; Heeger, A. J. Science 1995, 270, 1789–1791. 2. (a) Thompson, B. C.; Fre´chet, J. M. J. Angew
Chem Int Ed 2008, 47, 58–77; (b) Kietzke, T.; Shin, R. Y. C.; Ayuk Mbi Egbe, D.; Chen, Z. K.; Sellinger, A. Macromolecules 2007, 40, 4424– 4428; (c) Uhrich, C.; Schueppel, R.; Petrich, A.; Pfeiffer, M.; Leo, K.; Brier, E.; Kilickiran, P.; Baeuerle, P. Adv Funct Mater 2007, 17, 2991– 2999.
3. (a) Boucle´, J.; Ravirajanac, P.; Nelson, J. J Mater Chem 2007, 17, 3141–3153; (b) Beek, W. J. E.; Wienk, M. M.; Janssen, R. A. J. Adv Funct Mater 2006, 16, 1112–1116; (c) Moet, D. J. D.; Koster, L. J. A.; de Boer, B.; Blom, P. W. M. Chem Mater 2007, 19, 5856–5861; (d) Beek, W. J. E.; Wienk, M. M.; Janssen, R. A. J. Adv Mater 2004, 16, 1009–1013.
4. (a) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Mor-iarty, T.; Emery, K.; Yang, Y. Nat Mater 2005, 4, 864–868; (b) Koppe, M.; Scharber, M.; Brabec, C.; Duffy, W.; Heeney, M.; McCulloch, I. Adv Funct Mater 2007, 17, 1371–1376; (c) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv Funct Mater 2005, 15, 1617–1622.
5. (a) Coffey, D. C.; Reid, O. G.; Rodovsky, D. B.; Bartholomew, G. P.; Ginger, D. S. Nano Lett 2007, 7, 738–744; (b) Wienk, M. M.; Wiljan, J. M. K.; Verhees, J. H.; Knol, J.; Hummelen, J. C.; Hal, P.
A. v.; Janssen, R. A. J. Angew Chem Int Ed 2003, 42, 3371–3375.
6. (a) Alstrup, J.; Norrman, K.; Jorgensen, M.; Krebs, F. C. Sol Energy Mater 2006, 90, 2777– 2792; (b) Chabinyc, M. L.; Street, R. A.; Northrup, J. E. Appl Phys Lett 2007, 90, 123508–123513. 7. (a) Gadisa, A.; Mammo, W.; Andersson, L. M.;
Admassie, S.; Zhang, F.; Andersson, M. R.; Inga-na¨s, O. Adv Funct Mater 2007, 17, 3836–3842; (b) Ashraf, R. S.; Hoppe, H.; Shahid, M.; Gobsch, G.; Sensfuss, S.; Klemm, E. J Polym Sci Part A: Polym Chem 2006, 44, 6952–6961; (c) Li, K. C.; Hsu, Y. C.; Lin, J. T.; Yang, C. C.; Wei, K. W.; Lin, H. C. J Polym Sci Part A: Polym Chem 2008, 46, 4285–4304; (d) Lee, J.; Jung, B. J.; Lee, S. K.; Lee, J. I.; Cho, H. J.; Shim, H. K. J Polym Sci Part A: Polym Chem 2005, 43, 1845–1857; (e) Tsami, A.; Yang, X. H.; Farrell, T.; Neher, D.; Holder, E. J Polym Sci Part A: Polym Chem 2008, 46, 7794–7808.
8. (a) Tang, W. H.; Kietzke, T.; Vemulamada, P.; Chen, Z. K. J Polym Sci Part A: Polym Chem 2007, 45, 5266–5276; (b) Zhao, C. C.; Chen, X. H.; Zhang, Y.; Ng, M. K. J Polym Sci Part A: Polym Chem 2008, 46, 2680–2688; (c) Ayuk Mbi Egbe, D.; Nguyen, L. H.; Schmidtke, K.; Wild, A.; Sieber, C.; Guenes, S.; Sari-ciftci, N. S. J Polym Sci Part A: Polym Chem 2007, 45, 1619–1631; (d) Cheng, K. F.; Chueh, C. C.; Lin, C. H.; Chen, W. C. J Polym Sci Part A: Polym Chem 2008, 46, 6305–6316; (e) Ouhib, F.; Dkhissi, A.; Irat-c¸abal, P.; Hiorns, R. C.; Khoukh, A.; Desbrie`res, J.; Pouchan, C.; Dagron-Lartigau, C. J Polym Sci Part A: Polym Chem 2008, 46, 7505–7516.
9. (a) Xia, P. F.; Lu, J. P.; Kwok, C. H.; Fukutani, H.; Wong, M. S.; Tao, Y. J Polym Sci Part A: Polym Chem 2009, 47, 137–148; (b) Huo, L.; Tan, Z.; Wang, X.; Zhou, Y.; Han, M. F.; Li, Y. F. J Polym Sci Part A: Polym Chem 2008, 46, 4038–4049; (c) Huo, L.; He, H.; Han, M. F.; Zhou, E.; Li, Y. F. J Polym Sci Part A: Polym Chem 2007, 45, 3861– 3871.
10. (a) Scharber, M. C.; Mu¨hlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.; Brabec, C. J. Adv Mater 2006, 18, 789–794; (b) Gu¨nes, S.; Neu-gebauer, H.; Sariciftci, N. S. Chem Rev 2007, 107, 1324–1338; (c) Brabec, C. J.; Cravino, A.; Meiss-ner, D.; Sariciftci, N. S.; Fromherz, T.; Rispens, M. T.; Sanchez, L.; Hummelen, J. C. Adv Funct Mater 2001, 11, 374–380.
11. (a) Coppo, P.; Turner, M. L. J Mater Chem 2005, 15, 1123–1133; (b) Yang, J. S.; Liau, K. L.; Tu, C. W.; Hwang, C. Y. J. Phys Chem A 2005, 109, 6450–6456; (c) Pal, B.; Yen, W. C.; Yang, J. S.; Chao, C. Y.; Hung, Y. C.; Lin, S. T.; Chuang, C. H.; Chen, C. W.; Su, W. F. Macromolecules 2008, 41, 6664–6671.
12. (a) Kraak, A.; Wieserma, A. K.; Jordens, P.; Wyn-berg, H. Tetrahedron 1968, 24, 3381–3398; (b) Berlin, A.; Brenna, E.; Pagani, G. A.; Sannicolo,
F. Synth Met 1992, 51, 287–297; (c) Cunningham, D. D.; Gaa, A.; Pham, C. V.; Lewis, E. T.; Bur-khardt, A.; Davidson, L. L.; Kansah, A. N.; Ata-man, O. Y.; Zimmer, H.; Mark, H. B. J Electro-chem Soc 1988, 135, 2750.
13. (a) Zotti, G.; Schiavon, G. Macromolecules 1994, 27, 1938–1942; (b) Asawapirom, U.; Scherf, U. Macromol Rapid Commun 2001, 22, 746–749; (c) Coppo, P.; Cupertino, D. C.; Yeates, S. G.; Turner, M. L. Macromolecules 2003, 36, 2705–2711; (d) Cremer, L. D.; Vandeleene, S.; Maesen, M.; Verb-iest, T.; Koeckelberghs, G. Macromolecules 2008, 41, 591–598.
14. (a) Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J. Y.; Lee, K.; Bazan, G. C.; Heeger, A. J. J Am Chem Soc 2008, 130, 3619– 3623; (b) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G. C. Nat Mater 2007, 6, 497–500; (c) Mu¨1hlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C. Adv Mater 2006, 18, 2884–2889; (d) Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Mu¨1hlbacher, D.; Scharber, M.; Bra-bec, C. Macromolecules 2007, 40, 1981–1986; (e) Moule´, A. J.; Tsami, A.; Bu¨nnagel, T. W.; Forster, M.; Kronenberg, N. M.; Scharber, M.; Koppe, M.; Morana, M.; Brabec, C. J.; Meerholz, K.; Scherf, U. Chem Mater 2008, 20, 4045–4050; (f) Xiao, S.; Zhou, H.; You, W. Macromolecules 2008, 41, 5688– 5696.
15. (a) Zhan, X.; Tan, Z.; Domercq, B.; An, Z.; Zhang, X.; Barlow, S.; Li, Y.; Zhu, D.; Kippelen, B.; Marder, S. R. J Am Chem Soc 2007, 129, 7246– 7247; (b) Zhang, M.; Tsao, H. N.; Pisula, W.; Yang, C.; Mishra, A. K.; Mu¨llen, K. J Am Chem Soc 2007, 129, 3472–3473.
16. (a) Lambert, T. L.; Ferraris, J. P. J Chem Soc Chem Commun 1991, 752; (b) Ferraris, J. P.; Lambert, T. L. J. Chem Soc Chem Commun 1991, 1268; (c) Berlin, A.; Zotti, G.; Zecchin, S.; Schia-von, G.; Vercelli, B.; Zanelli, A. Chem Mater 2004, 16, 3667–3676.
17. (a) Roncali, J. Macromo Rapid Commun 2007, 28, 1761–1775; (b) van Mullekom, H. A. M.; Veke-mans, J. A. J. M.; Havinga, E. E.; Meijer, E. W. Mater Sci Eng 2001, 32, 1–40.
18. (a) Lucas, P.; El Mehdi, N.; Anh Ho, H.; Belanger, D.; Breau, L. Synthesis 2000, 9, 1253–1258; (b) Brzezinski, J. Z.; Reynolds, J. R. Synthesis 2002, 8, 1053–1056.
19. (a) Brabec, C. J.; Sariciftci, N. S.; Hummelen, J. C. Adv Funct Mater 2001, 11, 15–26; (b) Al-Ibra-hima, M.; Rotha, H. K.; Zhokhavetsb, U. Solar Energy Mater Solar Cells 2005, 85, 13–20.
20. (a) Wu, C. W.; Tsai, C. M.; Lin, H. C. Macromole-cules 2006, 39, 4298–4305; (b) Wu, C. W.; Lin, H. C. Macromolecules 2006, 39, 7232–7240.
21. (a) Liao, L.; Dai, L.; Smith, A.; Durstock, M.; Lu, J.; Ding, J.; Tao, Y. Macromolecules 2007, 40, 9406–9412; (b) Zhu, Y.; Champion, R. D.; Jenekhe, S. A. Macromolecules 2006, 39, 8712–8719. 22. (a) Osaka, I.; Sauve´, G.; Zhang, R.; Kowalewski,
T.; McCullough, R. D. Adv Mater 2007, 19, 4160– 4165; (b) Hou, J.; Tan, Z.; Yan, Y.; He, Y.; Yang, C.; Li, Y. J Am Chem Soc 2006, 128, 4911–4916. 23. (a) Hou, Y.; Chen, Y.; Liu, Q.; Yang, M.; Wan, X.;
Yin, S.; Yu, A. Macromolecules 2008, 41, 3114– 3119; (b) Colladet, K.; Fourier, S.; Cleij, T. J.; Lut-sen, L.; Gelan, J.; Vanderzande, D. Macromole-cules 2007, 40, 65–72.
24. Shahid, M.; Ashraf, R. S.; Klemm, E.; Sensfuss, S. Macromolecules 2006, 39, 7844–7853.
25. Leeuw, D. M.; Simenon, M. M. J.; Brown, A. R.; Einhard, R. E. F. Synth Met 1997, 87, 53–59. 26. (a) Roncali, J. J Mater Chem 1999, 9, 1875; (b)
Roncali, J. Chem Rev 1997, 97, 173–206.
27. Mihailetchi, V. D.; Duren, J. K. J.; Blom, P. W. M.; Hummelen, J. C.; Janssen, R. A. J.; Kroon, J. M.; Rispens, M. T.; Verhees, W. J. H.; Wienk, M. M. Adv Funct Mater 2003, 13, 43–46.
28. (a) Lu, G.; Usta, H.; Risko, C.; Wang, L.; Fac-chetti, A.; Ratner, M. A.; Marks, T. J. J Am Chem Soc 2008, 130, 7670–7685; (b) Yasuda, T.; Sakai, Y.; Aramaki, S.; Yamamoto, T. Chem Mater 2005, 17, 6060–6068.
29. (a) Yamamoto, T.; Arai, M.; Kokubo, H.; Sasaki, S. Macromolecules 2003, 36, 7986–7993.
30. (a) Cao, J.; Kampf, J. W.; Curtis, M. D. Chem Mater 2003, 15, 404–411; (b) Yamamoto, Y.; Lee, B. L. Macromolecules 2002, 35, 2993–2999; (c) Ong, B. S.; Wu, Y. L.; Liu, P.; Gardner, S. J Am Chem Soc 2004, 126, 3378–3379; (d) Morana, M.; Weg-scheider, M.; Bonanni, A.; Kopidakis, N.; Shaheen, S.; Scharber, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C. Adv Funct Mater 2008, 18, 1757– 1766.
31. (a) Morteani, A. C.; Sreearunothai, P.; Herz, L. M.; Friend, R. H.; Silva, C. Phys Rev Lett 2004, 92, 247402–247404; (b) Blom, P. W. M.; deJong, M. J. M.; vanMunster, M. G. Phys Rev B 1997, 55, R656; (c) Dunlap, D. H.; Parris, P. E.; Kenkre, V. M. Phys Rev Lett 1996, 77, 542–545.
32. Moule´, A. J.; Bonekamp, J. B.; Meerholz, K. J Appl Phys 2006, 100, 094503–094507.
33. (a) Markov, D. E.; Hummelen, J. C.; Blom, P. W.; Sieval, A. B. Phys Rev B 2005, 72, 045216– 045225; (b) Kietzke, T.; Horhold, H.; Neher, D. Chem Mater 2005, 17, 6532–6537.
