Synthesis, photophysics, and photovoltaic properties of low-band gap conjugated polymers based on thieno[3,4-c]pyrrole-4,6-dione: a combined experimental and computational study

Abstract

Four novel donor–acceptor (D–A) alternating copolymers were designed and successfully synthesized by the palladium-catalyzed Stille coupling and Suzuki coupling reactions. Utilizing thieno[3,4-c]pyrrole-4,6-dione (TPD) as an acceptor comonomer coupled with dialkoxybithiophene or cyclopentadithiophene as the donor gave polymers PTBT and PTCT. Employing carbazole as the donor and the dithiophene-substituted TPD serving as the acceptor monomers yielded polymers PTC1 and PTC2. Owing to the various strengths of electronic coupling between the donors and the acceptor unit, the band gaps of these polymers can be adjusted from 1.57 to 1.90 eV. Due to the different electron-donor ability of dialkoxybithiophene, cyclopentadithiophene, and carbazole, the HOMO energy levels of polymers were tuned from −5.34 to −5.67 eV, while LUMO levels remained relatively unchanged. The theoretical calculations provided insight to the observed photophysical properties of these polymers. Theoretically estimated band gaps and oxidation potentials correlate well with the experimental data. Carrier mobility and photovoltaic properties of TPD polymers were also investigated for which 1.3% power conversion efficiency was obtained from a blend of PTCT:PC₇₁BM (1 : 2) bulk-heterojunction device.

---

1. Introduction

The emergence of solar cells utilizing the photovoltaic effect to generate electricity from solar energy gives an appealing solution to our growing energy needs and to our desire for protecting the environment. Compared with silicon-based solar cells, the low cost, high flexibility, and diversity of organic photovoltaic cells makes them one of the most promising photovoltaic technologies of the near future.^1–3 However, power conversion efficiencies (PCEs) of polymer solar cells (PSCs) should be improved to allow practical applications in large scale. High-performance, band-gap engineered low-band gap polymers is a pre-requisite to achieve efficient harvesting of solar irradiation and charge separation when the donor polymer is mixed with the fullerene acceptors. Consequently, low-band gap PSCs have been investigated extensively in the past decade for the great potential to fabricate them onto large areas of lightweight flexible substrates by solution processing techniques.^4–11

Low-band gap polymers can be synthesized by coupling an electron-rich donor (D) with an electron-deficient acceptor (A) in the repeating unit of an alternating copolymer. Recently, Watson¹² and Leclerc¹³ successfully employed phthalimide and thieno[3,4-c]pyrrole-4,6-dione (TPD) as acceptor in donor–acceptor (D–A) type conjugated copolymers with good performances in organic electronic device applications. The reported D–A copolymer PhBT12 based on phthalimide had a narrow optical band gap at 1.64 eV and its thin films exhibited a high field-effect hole mobility of 0.28 cm² V⁻¹ s⁻¹ and a PCE of 2.0% of in a bulk heterojunction (BHJ) solar cell. Replacing the six-membered benzene ring in phthalimide with the five-membered ring of thiophene, TPD exhibited much better acceptor properties due to diminishing steric repulsion and aromatic resonance energy than phthalimide when incorporated into conjugated polymeric backbones. Therefore, TPD monomer has a great potential in building conjugated copolymers with strong intramolecular/intermolecular interactions, low optical band gaps, and low highest occupied molecular orbitals (HOMOs). A low-lying HOMO would not only increase the ambient stability of TPD-containing polymers but also would further increase the open-circuit voltage (V_oc) of photovoltaic devices. The copolymers based on TPD and benzodithiophene have shown excellent PCEs ranging from 4.0% to 6.8% as reported by various research groups.^14–17

In this study, we have designed and synthesized four novel D–A conjugated copolymers with TPD serving as acceptor comonomer. TPD was coupled to donors such as cyclopenta[2,1-b:3,4-b′]dithiophene (CPDT), carbazole, or 2,2′-dithiophene (Scheme 1). All these donor monomers have been previously exploited in high efficiency solar cells.^17–19Alkoxy substitution on thiophene groups is intended to adjust the energies of frontier levels further by introducing an extra electron-rich group into the conjugated backbone. This study aims to reveal how donors with various electron donating abilities affect the photophysical and photovoltaic properties of TPD-containing polymers. All polymers were fully characterized by gel-permeation chromatography (GPC), NMR, UV-vis absorption, photoluminescence (PL), and cyclic voltammetry (CV) methods. Photovoltaic properties are discussed along with an extensive computational study on the repeat units of the polymers. TPD-based copolymers are gaining interest and very recently, Guo et al. reported a D–A type polymer which incorporates TPD and CPDT in the backbone.²⁰

Scheme 1 Chemical structures of polymers. (R: 2-ethylhexyl.)

2. Results and discussion

2.1 Synthesis

The construction of D–A alternating copolymers lies in the efficient carbon–carbon single bond formation between the donor and the acceptor units. The most useful methods of forming C–C single bonds are based on transition metal-catalyzed cross-coupling reactions²¹ like Kumada–Corriu, Stille, Suzuki-Miyaura, and Sonogashira coupling reactions. PTCT and PTBT were synthesized by Stille coupling reaction with tris(dibenzylideneacetone)dipalladium(0) as the catalyst and tri(o-tolyl)phosphine as the ligand, while the Suzuki coupling reaction was utilized for the synthesis of PTC1 and PTC2 with the same catalyst and ligand (Scheme 2).

Scheme 2 Schematic procedure for the synthesis of polymers PTCT, PTBT, PTC1, and PTC2.

The heterocycle 1 was synthesized following the previously reported procedure by Brzeziński.²² Monomer 3 was obtained from 1 in two steps in a high yield. The 3-alkyl substituted bithiophene 6 being a regioregular system with head-to-tail type of linkage, could not be synthesized using Kumada–Corriu coupling reaction from 4 using nickel catalyst,²³ but was obtained-from a coupling reaction of 4 and 5 using a Pd(II) as catalyst. Purification with column chromatography afforded the product 6 which was found to be unstable in air and therefore, the next step was performed immediately. The two trimethylstannyl groups were introduced in the 5 and 5′ positions of 6 with the use of TMEDA activated n-butyl lithium. The low ambient stability of comonomer 7 also requires its immediate use in future synthetic steps. Monomer 9 was synthesized following the reported procedure²⁰ and was reacted with 3 or 7 respectively, to give polymer PTCT (82%) and PTBT (41%). The reactivity of comonomer 7 is less than that of 3, as evidenced from the low yield of polymerization. This can be explained with the large steric repulsion of alkoxy group at the 5′ position, slowing down the polymerization reaction. It is also possible for monomer 7 to lose the stannyl group with ease and result in unreactive chain ends during the course of polymerization.

Two thiophene-based units were easily introduced in TPD through Stille coupling reaction to give compounds 10 and 11. After bromination with NBS, 10 and 11 were smoothly converted into brominated monomers 12 and 13. Coupling them with monomer 14, PTC1 and PTC2 were obtained with the use of palladium catalyst. The lower yield of PTC2 (17%) compared to PTC1 (25%) was again attributed to the hindering effect of long dodecyloxyl chains during copolymerization.

Optical and electrochemical properties. The UV-vis absorption characteristics of the polymers were measured in both chloroform solution and in films (Fig. 1) with the relevant optical data collected in Table 1. The band gaps of the polymers vary between 1.57 eV and 1.90 eV. PTBT has the lowest band gap (1.57 eV) with a broad absorption profile covering most of the visible region. PTCT shows vibronic peaks both in solution and solid state with a band gap of 1.72 eV. PTC1 and PTC2 display larger band gaps and more importantly the main absorption band is around 500 nm. Although all of the polymers show a red-shift in the absorption spectra to some extent when going from solution to solid state, PTC1 displays the largest bathochromic shift along with a clear vibronic progression. Such vibronic signatures are usually indicative of ordering of polymer chains in a rigid film environment. Comparison of photoluminescence profiles both in the solution and in the film demonstrates that PTC1 has the largest red-shift in two different environments (Fig. S1, ESI†). Again, this can be explained by the presence of more planar conformations of the polymer chains adopted in the solid state.

Fig. 1 Normalized absorption spectra of PTCT, PTBT, PTC1, and PTC2 in chloroform solutions (top) and in thin films (bottom).

Table 1 Optical and electrochemical properties of the polymers along with the carrier mobilities measured with SCLC technique

Polymer	Solution		Film			Eox^onset (V)	HOMO (eV)	LUMO (eV)	μ (cm^2/(V s))
	λ max (nm)	PL (nm)	λ max (nm)	PL (nm)	Eg^opt (eV)
PTCT	652	684	670	727	1.72	1.05	−5.43	−3.71	6.2 × 10^−5
PTBT	583	669	584	806	1.57	0.96	−5.34	−3.77	6.1 × 10^−6
PTC1	482	528	534	689	1.90	1.29	−5.67	−3.77	1.2 × 10^−5
PTC2	484	566	482	724	1.71	1.09	−5.47	−3.76	1.2 × 10^−6

---

Since the acceptor chromophore is the same both in PTCT and PTBT, the lower band gap of PTBT shows stronger electron-donating capability of dialkoxybithiophene units relative to CPDT units, which is supported with the polymers' oxidation potentials as well as with the results obtained from theoretical calculations (see below). PTC2 exhibits a shoulder at 596 nm in solution and at 649 nm in film which probably corresponds to an intramolecular charge transfer state that involves dialkoxy bithiophene units which are absent in PTC1. As a result, PTC2 has a lower band gap than that of PTC1. This is expected since strong electron-donating alkoxy groups on thiophene rings push the HOMO higher; thus resulting in a lower band gap. However, the oscillator strength for the lowest excited state is relatively small; therefore the absorption profile very much resembles that of PTC1 except at higher wavelengths (Fig. 1, top).

Cyclic voltammetry is widely employed to examine the electrochemical properties and evaluate the frontier energy levels of conjugated polymers. The CV curves were obtained from the polymer films on a working electrode in a 0.1 M·Bu₄NPF₆ solution in acetonitrile. The redox couple ferrocene/ferrocenium ion (Fc/Fc+) was used as an external standard. The results are summarized in Table 1 and CV curves in the oxidation regime are shown in Fig. 2. These polymers exhibit reversible oxidation processes to varying degrees. The reversibility of oxidation in PTCT is better than other polymers. The oxidation potential (E_ox^onset) onsets were measured in the range from 0.96 to 1.29 eV vs.Ag/Ag⁺ couple. The onset of oxidation potential for PTBT is the lowest among the polymers studied in this work. When compared to PTCT, the low E_ox^onset of PTBT indicates that the dialkoxybithiophene donor possesses a stronger electron-donating ability than cyclopentadithiophene donor as we discussed above. HOMO and LUMO energy levels of the polymers were calculated according to the equation HOMO = −e(E_ox^onset − E_ox(Fc/Fc+)^onset + 4.8 V) and LUMO = HOMO + E_g^opt. Electrochemical data in Table 1 suggest that the LUMO is mostly determined by an electron-accepting unit, TPD, whereas the location of the HOMO level is determined by the strength of electron donating groups in the conjugated backbone. This is because the LUMO of the polymers slightly vary around −3.7 eV; on the other hand there is large breadth of HOMO energies varying from −5.34 to −5.67 eV. The fact that all of the polymers possess oxidation potentials larger than 5.2 eV suggests their stability at ambient conditions.

Fig. 2 Cyclic voltammograms of polymer thin films at a scan rate of 100 mV s⁻¹.

Carrier mobility and photovoltaic properties. The magnitudes of carrier mobilities in conjugated polymers usually determine the device performances of solar cells and thin film transistors. The polymers with high hole mobilities can be fabricated in thick films and at the same time enable fabrication of solar cells in roll-to-roll printing techniques. Besides, the thick films can increase the light harvesting efficiency of solar cells. Here, we employed space charge limited current (SCLC) technique to determine—presumably— the hole mobility of these polymers. The SCLC is described by^24,25

J = 9ε₀ε_rμV² L⁻³ (1)

where ε₀ is the permittivity of free space, ε_r is the dielectric constant of the medium, μ is the carrier mobility, V is the voltage drop across the device, and L is the thickness of the active layer. The SCLC results of devices (ITO/PEDOT:PSS/polymer/Al) are given in Table 1. The largest mobility measurement was obtained from a PTCT device (6.2 × 10⁻⁵ cm² V⁻¹ s⁻¹), though it is still lower than the hole mobility of well known polymer poly(3-hexylthiophene) (P3HT) (3 × 10⁻⁴ cm² V⁻¹ s⁻¹) measured under identical conditions.

Solar cells prepared from TPD polymers were studied, and representative characteristics are summarized in Table 2. All of the devices showed rather poor fill factors, hinting the non-optimal morphology for the active layers of blends. Relatively low molecular weights of the polymers could also lead to poor blend morphology which is detrimental for efficient photocurrent generation. As shown in Fig. 3A, the PTCT:PC₆₁BM device exhibited the best photovoltaic performance among the polymers studied under identical PC₆₁BM loading. The relatively high V_oc of PTCT could be attributed to the deeper HOMO level compared to that of P3HT. Therefore, we have also investigated the solar cell performance with the use of 1-chloronaphtalene³⁶ (CN) as a processing additive. By incorporating 2% CN into the blend film, PCEs of devices did not show significant changes. Addition of CN additive, however, slightly improved the fill factor (FF) in all devices. Nonetheless, the drop in short-circuit current (J_sc) undermined the improvement in photovoltaic properties due to fill factor enhancement (Table 2).

Fig. 3 (A) Current–voltage (I–V) characteristics of polymer/PC₆₁BM (1 : 1, w/w) solar cells. (B) I–V curves of the PTCT:PC₇₁BM (1 : 2, w/w) under light (a) and (b) dark conditions.

Table 2 Photovoltaic performance parameters for polymer:PC₆₁BM devices under simulated AM 1.5G illumination at 100 mW cm⁻²

Sample	Voc (V)	Jsc (mA cm^−2)	FF	PCE (%)
PTCT:PC61BM(1 : 1)	0.75	3.71	38.46	1.09
PTBT:PC61BM(1 : 1)	0.49	0.86	31.20	0.13
PTC1:PC61BM(1 : 1)	0.69	1.75	30.16	0.37
PTC2:PC61BM(1 : 1)	0.70	0.44	22.97	0.07
With 2% CN additive
PTCT:PC61BM(1 : 1)	0.78	3.02	39.68	1.11
PTBT:PC61BM(1 : 1)	0.53	0.67	37.82	0.13
PTC1:PC61BM(1 : 1)	POOR FILM	—	—	—
PTC2:PC61BM(1 : 1)	0.37	0.43	36.60	0.06

---

The best PCEs were measured in solar cells fabricated from PTCT polymer. The average PCE achieved from a PTCT:PC₆₁BM (1 : 1) blend with LiF/Al top electrode was 1.09% with J_sc = 3.71 mA cm⁻², V_oc = 0.75 V, and FF = 38.5. To further optimize the efficiency of PTCT-based devices, blend ratios (1 : 1 or 1 : 2), type of additives (CN or 1,8-diiodooctane), annealing temperature (50, 110, or 160 °C), top electrode (LiF/Al or Ca/Al), fullerene derivative (PC₆₁BM or PC₇₁BM) were varied and the highest efficiency achieved in a PTCT:PC₇₁BM (1 : 2) device with Ca/Al top electrode was 1.29%, J_sc = 4.49 mA cm⁻², V_oc = 0.80 V, and FF = 36.0 (Fig. 3B). The current–voltage curve of the PTCT:PC₇₁BM (1 : 2) device shows some S-curve behavior and this has been attributed to poor contact of the cathode with the PCBM phase, which hampers electron extraction and leads to an imbalance in the rates at which carriers are collected at each electrode.²⁶

Theoretical calculations. Density functional theory (DFT) calculations were performed on model molecules to predict geometrical features for the repeat units. To examine the torsional potential profile and also to determine the energy differences between different conformations, self-consistent field calculations were performed on isolated portions of the optimized repeat units by varying the twisting angle around the C–C single bond connecting the comonomers (Fig. S2, ESI†). Such a conformational study is necessary, as previous reports^27–28 claim that conjugated backbone planarization of TPD based polymers is due to potentially attractive sulfur–oxygen (S=O) interactions in the trans-configuration. However, our calculations on various conformers indicate that aromatic hydrogen on thiophene rings has a stronger interaction with the oxygen of the carbonyl oxygen, favoring a cis-geometry rather than a trans-configuration (see ESI† for detailed discussion).

Band gaps of donor–acceptor type alternating copolymers can be adjusted through careful selection of comonomers with different frontier energy levels. DFT calculated energy levels and frontier orbitals of comonomers and the repeat units of polymers are shown in Fig. 4. The TPD comonomer has a HOMO energy of −7.30 eV whereas the LUMO sits at −1.78 eV. Compared to a single thiophene ring, the HOMO is stabilized around 1 eV and the LUMO is stabilized around 1.5 eV. That means TPD has a lower HOMO–LUMO difference than thiophene and could serve as an electron acceptor when coupled to thiophene units in a polymer chain. In this work, TPD has been coupled to several donors and it has served as an electron deficient comonomer for band gap optimization. Since TPD has a deep HOMO level, the HOMO levels of repeat units mostly represent the HOMO of the donor comonomer. According to second order perturbation theory, the large differences in the energy levels of coupled units result in little interaction of each chromophore. This is usually the case when we analyze the frontier energy levels of repeat units in comparison to those of comonomers. For instance, in TC1 and TC2 repeat units the TPD comonomer determines the magnitude of the LUMO level. The LUMO of TPD mixes better with the LUMO of CPDT and dialkoxybithiophene due to relatively closer LUMO levels of interacting units. Although these are the results extracted with repeat units only, the CV data of polymers compare well with the changes in the energy levels of both HOMOs and LUMOs of repeat units.

Fig. 4 Comparison of frontier energy levels and HOMO/LUMO plots of TCT, TBT, TC1, and TC2 repeat units and their comonomers.

DFT-optimized geometries and HOMO/LUMO plots for two repeat units of polymers are given in Fig. 5. TCT and TBT dimers possess a totally flat conjugated backbone whereas TC1 and TC2 dimers show significant deviation from planarity. The sources of the deviation are carbazole linkages with the other segments in the repeat units, however the TPD-bithiophene unit still retains its planarity in TC1 and TC2 repeat units. All of the molecular orbital coefficients are in general well-delocalized over the structures except for those of the TBT dimer. As the size of the oligomer increases, however, the frontier orbitals localize in the conjugated backbone (Fig. S3, ESI†). The lowest excited state transition in TBT dimer occurs mostly between HOMO and LUMO orbitals (see Table S2, ESI†), hence this excited state has significant charge transfer character. On the other hand, the excited states of all the other oligomers are dominated by π–π* transitions.

Fig. 5 Optimized geometries and HOMO/LUMO plots of dimers based on synthesized polymers. Aliphatic carbon chains have been truncated to methyl groups. Optimized structures for TC1 and TC2 display significant backbone twisting while those for TCT and TBT are totally planar. Differences in HOMO/LUMO localization are found to varying degrees, but are especially evident in the TBT dimer.

Geometry optimizations were also performed with BHandH/6-31G(d) for neutral and cationic systems in the monomer to tetramer series, beginning with ground state geometries calculated using the B3LYP hybrid functional. We previously showed that fitting the phenomenological function and parameters of Meier²⁹ when extrapolating the theoretical band gap and calculated oxidation potentials gives reasonably accurate results when compared with the experiment.³⁰ In our prior study, it was also found that the BHandH functional had the most accurate performance when estimating the oxidation potential (scaled down by 0.2 eV) in comparison with the results obtained from other commonly used DFT functionals. For time-dependent density functional theory (TD-DFT) calculations, we have used the B3LYP/6-31G(d) method to calculate the excited state transition energies for monomers through tetramer oligomers (Fig. 6). The oxidation potential was calculated according to the reaction;

M(g) → M⁺(g) + e⁻ (2)

Fig. 6 (A) Calculated band gaps of oligomers from TD-DFT calculations using B3LYP/6-31G(d) optimized ground state geometries and (B) theoretical oxidation potentials estimated with the BHandH functional. Included are the corresponding parameterized Meier fits for increasing oligomer chain length. Theoretically estimated values are compared with the experimental values in the insets. Experimental band gaps were taken from solution spectra. Note that calculated oxidation potentials have been scaled down by 0.2 eV based on the work in ref. 30.

Plots of calculated band gap or oxidation potential were then fit with Meier’s equation to estimate the converged values at the infinite chain length limit;

where OP¹(E¹_g) is the oxidation potential (band gap) of the parent oligomer, OP^∞(E^∞_g) is the oxidation potential (band gap at infinite chain length), n is number of repeat units, and the parameter a describes the rate at which the OP (E_g) stabilizes with additional repeat units. The zero-point energy of monomer and dimer oligomers were also determined for oxidation potential calculations, but were on the order of 0.01 eV and were neglected due to little improvement in accuracy of the calculation and expensive computational cost. Extrapolated values for band gap and oxidation potential to infinite chain length limit are displayed in the insets of Fig. 6. The calculated band gaps and oxidation potentials in general agree well with the experimental values, capturing the trend observed experimentally. Evidently, quantum mechanical calculations on target conjugated polymers can be quite useful if one utilizes the right computational method and data fitting procedure to predict the oxidation potentials and optical band gaps of conjugated polymers for photovoltaic applications before any synthetic attempts are made.

3. Conclusions

In summary, we have incorporated TPD as an electron-deficient unit in various D–A alternating copolymers. The band gaps of these polymer varied between 1.57 eV and 1.90 eV, with oxidation potential ranging from −5.34 to −5.67 eV. The best solar cells using PTCT as donor and PC₇₁BM as acceptor exhibit device performance with a V_oc of 0.80 V and a PCE of 1.3%. DFT calculations on oligomers predict E_g and OP of the polymers quite well after Meier fitting. Optimized geometries of the repeat units show that the oxygen atoms on TPD monomer forms a weak hydrogen bonding with the hydrogens on adjacent comonomer; thus planarizing the conjugated backbone. Introduction of carbazole group, however, causes deviation from planarity along the chain. The results point out the importance of theoretical calculations in predicting the photophysical properties of conjugated polymers for photovoltaic applications. Theory guided efforts could speed up the process for finding the suitable polymers for target synthesis.

4. Experimental section

General method and materials

High-resolution mass spectra were obtained using a Bruker Daltronics BioTOF HRMS spectrometer. NMR spectra were recorded in CDCl₃ on a Varian 300 MHz, 400 MHz or 500 MHz spectrometer, ¹H and ¹³C NMR spectra were reported in ppm relative to residual solvent. Multiplicity of all carbon signals recorded are singlet unless otherwise stated. All UV-vis spectra were taken as solutions in chloroform or as thin films spin-coated onto quartz substrates using a Varian Cary 50 spectrophotometer. Cyclic voltammetry measurements were carried out on an EDAQ Potentiostat 466 system. A conventional three-electrode configuration consisting of a glassy carbon working electrode, a Pt-wire counter electrode and an Ag/AgNO₃ reference electrode was used. The supporting electrolyte was [Bu₄N]PF₆ (0.1 M). Ferrocene was used as an external standard, and all potentials were quoted with reference to the ferrocene-ferrocenium (Fc/Fc⁺) couple at a scan rate of 100 mV s⁻¹. The oxidation potential (E_ox) was used to determine the HOMO energy level using the equation E_HOMO = −(E_ox + 4.8) eV, which was calculated to the value of ferrocene (−4.8 eV).^31,32GPC analysis utilized a Symyx Rapid spectrometer with an Agilent 1110 Isocratic pump, PL-ELS 1000 Evaporative Ligh Scattering Detector and Varian PL-gel columns (1110–1120 and 1110–6100). THF was used as the solvent for GPC with a flow rate of 2.0 cm³ min⁻¹ and the calibration curve was made with a series of monodisperse polystyrene standards (580–377 400 M_w) from EasiCal Varian. Emission spectra were recorded on a Jobin-Yvon Fluorolog 3 fluorescence spectrometer.

Tetrahydrofuran was dried over sodium and freshly distilled prior to use. All other chemicals were purchased from Sigma-Aldrich and were used as received. Precoated thin-layer chromatographic plates (TLC) and silica gel (230–400 mesh) were purchased from Sorbent. Compound 1–3,³³4,²³8,¹²9,¹³ and 14³⁴ were synthesized according to reported procedures.

Tributyl(4-(dodecyloxy)thiophen-2-yl)stannane (5)

A 2.0 M solution of lithium diisopropylamide (12.8 mL, 25.6 mmol) in ethylbenzene/THF/heptane was added dropwise to a solution of 3-dodecyloxythiophene (6.85 g, 25.5 mmol) in 100 ml dry THF at −78 °C. The mixture was stirred at −78 °C for 2 h. Then, tributyltin chloride (8.0 mL 29.5 mmol) was added slowly. The solution was stirred for an additional 2 h at −78 °C and then overnight at room temperature. Water was then added under vigorous stirring. The aqueous layer was removed, and the organic layer was washed with distilled water. The organic layer was dried over magnesium sulfate and the solvent was removed under reduced pressure. The crude product was used directly without further purification.

3,4′-Bis(dodecyloxy)-2,2′-bithiophene (6)

In a 100 mL flask fitted with a condenser, compound 4 (2.8 g, 8 mmol), 5 (6.7 g, 12.0 mmol), and 50 mL DMF were added. Then, bis(triphenylphosphine)palladium(II) dichloride (260 mg, 0.37 mmol) was added and the reaction mixture was heated to 90 °C and stirred overnight. The reaction mixture was diluted with dichloromethane and washed with water and dried over MgSO₄. After concentration under reduced pressure, the residue purified by column chromatography on silica gel (DCM:hexane = 1 : 7) to give a yellow solid (2.8 g, 65%). ¹H NMR (CD₃COCD₃, 400 MHz): δ 0.85–0.91 (6H, m), 1.67–1.80 (36H, m), 1.80–2.06 (4H, m), 3.97 (2H, t, J = 6.4 Hz), 4.15 (2H, t, J = 6.4 Hz), 6.29 (1H, s), 6.88 (1H, d), 7.01 (1H, d, J = 5.6 Hz), 7.26 (1H, d, J = 5.6 Hz). ¹³C NMR (CD₃COCD₃, 100 MHz): δ 13.25, 13.69, 18.94, 22.65, 26.09, 26.77, 28.06, 28.59, 28.78, 28.99, 31.95, 69.88, 71.67, 95.44, 114.55, 117.95, 122.13, 129.05, 134.12, 153.45, 157.52. HRMS (ESI): m/z calcd. for C₃₂H₅₄O₂S₂Na (M⁺ + Na) 557.3457, found 557.3446.

(3,4′-Bis(dodecyloxy)-[2,2′-bithiophene]-5,5′-diyl)bis(trimethylstannane) (7)

Compound 6 (2.77 g, 5.2 mmol) was mixed with THF (20 mL) and TMEDA (1.56 g, 13 mmol) and cooled to −78 °C. A 2.5 M solution of n-BuLi (5.2 mL, 13 mmol) in hexane was added and the cooling bath was removed and the mixture was left overnight at RT. The mixture was cooled to −78 °C again and Me₃SnCl solution (13 mL, 13 mmol) in THF was added and the mixture was allowed to reach RT. After stirring at room temperature for 1 h hexane (250 mL) was added and the mixture was washed thoroughly with water and brine. Drying with MgSO₄ and evaporation gave the product (4.3 g, 96%). ¹H NMR (CD₃COCD₃, 500 MHz): δ 0.34–0.37 (18H, m), 0.87–0.91 (6H, m), 1.29–1.36 (36H, m), 1.51–1.53 (2H, m), 1.77–1.79 (2H, m), 4.02 (2H, t, J = 6.3 Hz), 4.18 (2H, t, J = 6.4 Hz), 7.10 (1H, s), 7.19 (1H, s). ¹³C NMR (CD₃COCD₃, 400 MHz): δ −9.02, 13.30, 13.71, 22.67, 26.13, 26.40, 27.37, 28.57, 28.96, 29.34, 29.65, 29.75, 31.98, 45.39, 71.49, 71.60, 112.07, 125.36, 134.52, 140.63, 154.65, 163.84. HRMS (ESI): m/z calcd for C₃₈H₇₀O₂S₂Sn₂Na (M⁺ + Na) 885.2764, found 885.2747.

5-Dodecyl-1,3-di(thiophen-2-yl)-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (10)

In a 100 mL flask fitted with a condenser, 9 (1.43 g, 3.0 mmol), 8 (1.95 mL, 6.1 mmol) and 20 mL DMF were added. Then, bis(triphenylphosphine)palladium(II) dichloride (106 mg, 0.15 mmol) was add and the reaction mixture was heated to 90 °C and stirred overnight. The reaction mixture was cooled to room temperature and the solvent was evaporated. The crude product was purified by column chromatography on silica gel (DCM:hexane = 1 : 3) to give a yellow solid (1.36 g, 93%). ¹H NMR (CDCl₃, 400 MHz): δ 0.85–0.94 (3H, m), 1.24–1.35 (18H, m), 1.62–1.67 (2H, m), 3.61 (2H, t, J = 7.2 Hz), 7.05 (2H, dd, J = 4.0, 3.6 Hz), 7.37 (2H, d, J = 4.4 Hz), 7.95 (2H, d, J = 3.6 Hz). ¹³C NMR (CDCl₃, 100 MHz): δ 14.34, 22.90, 27.19, 28.70, 29.46, 29.57, 29.73, 29.81, 29.84, 32.13, 38.74, 128.55, 128.60, 128.73, 130.02, 132.65, 136.51, 162.62. HRMS (ESI): m/z calcd. for C₂₆H₃₁NO₂S₃Na (M⁺ + Na) 508.1409, found 508.1405.

5-Dodecyl-1,3-bis(4-(dodecyloxy)thiophen-2-yl)-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (11)

The compound 11 was synthesized using the same procedure for 10, with 9 (2.0 g, 4.2 mmol), 5 (7.0 g, 12.6 mmol), 30 mL DMF and bis(triphenylphosphine)palladium(II) dichloride (148 mg, 0.2 mmol). Purification by column chromatography gave 11 as a red solid (3.1 g, 86%). ¹H NMR (CDCl₃, 400 MHz): δ 0.86–0.90 (9H, m), 1.25–1.32 (54H, m), 1.76–1.80 (6H, m), 3.64 (2H, t, J = 7.2 Hz), 3.97 (4H, J = 6.4 Hz), 6.37 (2H, s), 7.68 (2H, s). ¹³C NMR (CDCl₃, 100 MHz): δ 14.33, 22.89, 26.01, 26.19, 27.17, 28.69, 29.42, 29.58, 29.70, 29.78, 29.81, 29.88, 32.13, 38.83, 70.69, 101.44, 121.50, 128.83, 130.83, 136.87, 158.29, 162.70. HRMS (ESI): m/z calcd. for C₅₀H₇₉NO₄S₃Na (M⁺ + Na) 876.5063, found 876.5044.

1,3-Bis(5-bromothiophen-2-yl)-5-dodecyl-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (12)

NBS (0.74 g, 4.2 mmol) was added in one portion to 10 (0.97 g, 2.0 mmol) in 8 mL DMF at 0 °C. The mixture was warmed to room temperature and stirred for 24 h. Then the reaction mixture was diluted with ether (20 mL) and washed with water (2 × 20 mL). The organic layer was dried over MgSO₄, concentrated under reduced pressure, and the crude product was purified by column chromatography on silica gel (DCM:hexane = 1 : 3) to give 12 as a yellow solid (1.1 g, 85%). ¹H NMR (CDCl₃, 400 MHz): δ 0.87 (3H, t, J = 6.8 Hz), 1.25–1.32 (18H, m), 1.59–1.68 (2H, m), 3.58 (2H, t, J = 7.2 Hz), 6.99 (2H, d, J = 4.0 Hz), 7.56 (2H, d, J = 4.0 Hz). ¹³C NMR (CDCl₃, 100 MHz): δ 14.34, 22.89, 27.17, 28.65, 29.42, 29.57, 29.71, 29.80, 29.84, 32.13, 38.85, 116.94, 128.73, 129.90, 131.29, 133.92, 135.12, 162.34. HRMS (ESI): m/z calcd. for C₂₆H₂₉Br₂NO₂S₃Na (M⁺ + Na) 663.9619, found 663.9620.

1,3-Bis(5-bromo-4-(dodecyloxy)thiophen-2-yl)-5-dodecyl-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (13)

The compound 13 was synthesized using the similar procedure for 12 with 11 (2.0 g, 2.3 mmol), NBS (0.85 g, 4.8 mmol) and 8 mL DMF, except that the mixture was stirred at 60 °C for 24 h. The crude product was purified by column chromatography to give 13 as a brown solid (1.2 g, 53%). ¹H NMR (CDCl₃, 400 MHz): δ 0.85–0.90 (9H, m), 1.25–1.33 (54H, m), 1.62–1.68 (2H, m), 1.73–1.80 (4H, m), 3.57 (2H, t, J = 7.2 Hz), 4.03 (4H, t, J = 6.4 Hz), 7.61 (2H, s). ¹³C NMR (CDCl₃, 100 MHz): δ 14.34, 22.92, 26.10, 27.29, 28.77, 29.53, 29.59, 29.66, 29.85, 29.88, 29.94, 32.16, 38.94, 72.52, 95.56, 118.72, 128.54, 130.21, 135.52, 155.40, 162.28. HRMS (ESI): m/z calcd. for C₅₀H₇₇Br₂NO₄S₃Na (M⁺ + Na) 1034.3258, found 1034.3241.

General procedure for the syntheses of polymers by Stille coupling reaction

0.5 mmol of dibromide (9) and 0.5 mmol of bis(trimethylstannane) (3 or 7) were dissolved in 10 mL of toluene, and then Pd₂dba₃ (18.3 mg, 0.02 mmol) and P(o-Tolyl)₃ (65.44 mg, 0.16 mmol) were added into the flask. The mixture was flushed with nitrogen for 10 min. The polymerization reaction flask was heated to 110 °C, and the mixture was stirred for 48 h under nitrogen atmosphere. 2-tributylstannyl thiophene (0.06 mL, 0.2 mmol) was added to the reaction and then after 6 h, 2-bromothiophene (0.02 mL, 0.2 mmol) was added. The mixture was stirred overnight to complete the end-capping reaction. The mixture was cooled to room temperature and poured slowly in methanol (400 mL). The solid was filtered through a 0.45 μm nylon filter. The crude polymer was washed with acetone and hexanes in a soxhlet apparatus to remove the oligomers and catalyst residue. Finally the polymer was extracted with chloroform. The polymer solution was condensed to about 20 mL and slowly poured in methanol (400 mL). The solid was obtained after filtration.

PTCT

Compound 9 (0.24 g, 0.5 mmol) and 3 (0.37 g, 0.5 mmol) were used as comonomers. PTCT was obtained as dark blue solid (0.30 g, 82%). GPC: M_n, 15 100 Da; M_w, 20380 Da; PDI, 1.35. ¹H NMR (400 MHz, CDCl₃, δ (ppm)): 0.64–0.75 (m, 12H), 0.98 (m, 17H), 1.27 (m, 22H), 1.68 (m, 2H), 2.01 (br, 4H), 3.71 (br, 2H), 7.99 (m, 2H).

PTBT

Compound 9 (0.24 g, 0.5 mmol) and 7 (0.43 g, 0.5 mmol) were used as comonomers. PTBT was obtained as dark blue solid (0.18 g, 41%). GPC: M_n, 5859 Da ; M_w, 8504 Da; PDI, 1.45. ¹H NMR (400 MHz, CDCl₃, δ (ppm)): 0.84 (br, 9H), 1.23 (br, 48H), 1.61–1.99 (m, 12H), 3.60 (m, 2H), 4.16 (m, 4H), 6.80 (br, 1H), 7.41 (m, 1H).

General procedure for the syntheses of polymers by Suzuki coupling reaction

0.5 mmol of dibromide (12 or 13), 0.5 mmol of 9-(2-ethylhexyl)-2,7-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-9H-carbazole (14), Pd₂dba₃ (18.3 mg, 0.02 mmol) and P(o-Tolyl)₃ (65.44 mg, 0.16 mmol) were dissolved in 12.0 mL of toluene and 2.5 mL of 2N·Na₂CO₃ aqueous. The mixture was flushed with nitrogen for 10 min and vigorously stirred at 100 °C. After 48 h, bromobenzene (0.02 mL, 0.2 mmol) was added to the reaction then 6 h later, phenylboronic acid (24.4 mg, 0.2 mmol) was added and the reaction refluxed overnight to complete the end-capping reaction. The polymer was purified by precipitation in methanol/water (10 : 1), filtered through 0.45 μm nylon filter and washed on soxhlet apparatus with acetone, hexanes and chloroform. The chloroform fraction was reduced to 20 mL under reduced pressure, precipitated in methanol (400 mL), filtered through 0.45 lm nylon filter and finally air dried overnight.

PTC1

Compound 14 (0.27 g, 0.5 mmol) and 12 (0.32 g, 0.5 mmol) were used as comonomers. PTC1 was obtained as dark red solid (0.10 g, 25%). GPC:M_n, 5967 Da; M_w, 11 040 Da; PDI, 1.85. ¹H NMR (400 MHz, CDCl₃, δ (ppm)): 0.87 (m, 9H), 1.26 (m, 25H), 1.71 (m, 4H), 3.68 (m, 2H), 4.07 (m, 2H), 7.17 (m, 2H), 7.34–7.68 (m, 6H), 8.03 (m, 2H).

PTC2

Compound 14 (0.27 g, 0.5 mmol) and 13 (0.51 g, 0.5 mmol) were used as comonomers. PTC2 was obtained as brown solid (0.07 g, 17%). GPC: M_n, 3761 Da; M_w, 5248 Da; PDI, 1.40. ¹H·NMR (400 MHz, CDCl₃, δ (ppm)): 0.87 (m, 15H), 1.26 (m, 57H), 1.50–1.96 (m, 12H), 3.57 (m, 2H), 3.95–4.15 (m, 6H), 7.23–7.64 (m, 4H), 7.77–7.96 (m, 4H).

Device fabrication

Photovoltaic properties of the polymers were investigated by solar cell structures of ITO/PEDOT:PSS/polymer:PC₆₁BM(1 : 1, w/w)/LiF/Al and ITO/PEDOT:PSS/PTCT:PC₇₁BM(1 : 2, w/w)/Ca/Al. The PSCs were fabricated in the configuration of the traditional sandwich structure with an indium tin oxide (ITO) anode and a metal top-electrode. Patterned ITO glasses with a sheet resistance of 10 Ω/sq were purchased from Thin Film Devices, Inc. The ITO glass was cleaned by sequential ultrasonic treatment in detergent, deionized water, acetone, and isopropanol, and then treated in a bench-top plasma cleaner (PE-50 benchtop cleaner, The Plasma Etch, Inc., USA) for 2 min. Then, PEDOT:PSS (poly (3,4-ethylenedioxythiophene):poly(styrene sulfonate)) (Clevious P VP AI 4083 H. C. Stark, Germany) was filtered through a 0.45 μm filter and spin coated at 4000 rpm for 60 s on the ITO electrode. Subsequently, the PEDOT:PSS film was baked at 120 °C for 40 min in the air. After transferring to a N₂ filled glove box, the blend solution of polymer and fullerene derivative in chlorobenzene (CB) (1 : 1 w/w, 15 mg mL⁻¹) was spin-coated on top of the PEDOT:PSS layer at 600 rpm for 30 s. The wet polymer:PCBM blend films were then put into glass petri dishes to undergo solvent aging process for 60 min. The active layer was annealed at 50 °C for 10 min on the hotplate in the glovebox. Then, the cathode consisting of Ca (∼20 nm) capped with Al (∼80 nm) or LiF (∼0.6 nm) capped with Al (∼80 nm) was thermally evaporated on the active layer under a shadow mask in a base pressure of 1 × 10⁻⁶ mbar. The polymer photovoltaic cells were encapsulated in the glove box with a UV-curable epoxy under glass sheets and then taken out for current–voltage (I–V) measurements. The device active area was ∼7.9 mm² for all the solar cells discussed in this work. The I–V measurement of the devices was conducted on a computer controlled Keithley 2400 source meter. The I–V measurement system uses a solar simulator with a Class-A match to the AM1.5 Global Reference Spectrum. It is calibrated with KG5-filtered silicon reference cell with calibration traceable to NREL and NIST. Film thicknesses of polymer films were obtained with a Veeco (Dektak) step profiler for SCLC measurements.

Computational methods

Density functional theory calculations were performed on oligomeric analogs of the polymers synthesized in this study using the Gaussian 03 and Gaussian 09 software packages.³⁵ Optimizations for oligomers were performed using the Becke hybrid functional for exchange and Lee, Yang and Par functional for correlation (B3LYP) with a 6-31G(d) basis set. All alkyl groups were truncated to methyl groups to save computational time. Comonomer and repeat units were optimized without any symmetry restrictions after a rigorous conformational study on torsional potentials. Higher order oligomers were formed from the lowest energy repeat unit geometries following a similar torsional potential study on the bonds connecting the repeat units. Although several other possible conformations were optimized, the lowest energy conformations proved to be a conglomerate of the lowest energy repeat unit geometries.

Acknowledgements

This work was supported by ND EPSCoR, Department of Energy (DOE) EPSCoR seed grant program (SUNRISE Program), and DOE under award #DE-FG52-08NA28921.

References

1. C. W. Tang, Appl. Phys. Lett., 1986, 48, 183.
2. N. S. Sariciftci, D. Braun, C. Zhang, V. I. Srdanov, A. J. Heeger, G. Stucky and F. Wudl, Appl. Phys. Lett., 1993, 62, 585.
3. M. Hiramoto, H. Fujiwara and M. Yokoyama, Appl. Phys. Lett., 1991, 58, 1062.
4. C. J. Brabec, N. S. Sariciftci and J. C. Hummelen, Adv. Funct. Mater., 2001, 11, 15.
5. K. M. Coakley and M. D. McGehee, Chem. Mater., 2004, 16, 4533.
6. Y. J. Cheng, S. H. Yang and C. S. Hsu, Chem. Rev., 2009, 109, 5868.
7. S. Gunes, H. Neugebauer and N. S. Sariciftci, Chem. Rev., 2007, 107, 1324.
8. H. Hoppe and N. S. Sariciftci, J. Mater. Res., 2004, 19, 1924.
9. B. C. Thompson and J. M. J. Fréchet, Angew. Chem., Int. Ed., 2008, 47, 58.
10. B. C. Thompson and J. M. J. Fréchet, Angew. Chem., Int. Ed., 2008, 47, 58.
11. S. P. Xiao, H. Zhou and W. You, A. C. S. Symp. Ser., 2010, 71, 1034.
12. X. G. Guo, F. S. Kim, S. A. Jenekhe and M. D. Watson, J. Am. Chem. Soc., 2009, 131, 7206.
13. Y. P. Zou, A. Najari, P. Berrouard, S. Beaupre, B. R. Aich, Y. Tao and M. Leclerc, J. Am. Chem. Soc., 2010, 132, 5330.
14. Y. Zhang, S. K. Hau, H. L. Yip, Y. Sun, O. Acton and A. K. Y. Jen, Chem. Mater., 2010, 22, 2696.
15. C. Piliego, T. W. Holcombe, J. D. Douglas, C. H. Woo, P. M. Beaujuge and J. M. J. Fréchet, J. Am. Chem. Soc., 2010, 132, 7595.
16. G. B. Zhang, Y. Y. Fu, Q. Zhang and Z. Y. Xie, Chem. Commun., 2010, 46, 4997.
17. M. C. Yuan, M. Y. Chiu, S. P. Liu, C. M. Chen and K. H. Wei, Macromolecules, 2010, 43, 6936.
18. Z. Li, J. F. Ding, N. H. Song, J. P. Lu and Y. Tao, J. Am. Chem. Soc., 2010, 132, 13160.
19. C. J. Shi, Y. Yao, Y. Yang and Q. B. Pei, J. Am. Chem. Soc., 2006, 128, 8980.
20. X. Guo, H. Xin, F. S. Kim, A. D. T. Liyanage, S. A. Jenekhe and M. D. Watson, Macromolecules, 2010, 44, 269.
21. Y.-J. Cheng and T.-Y. Luh, J. Organomet. Chem., 2004, 689, 4137.
22. J. Z. Brzezinski and J. R. Reynolds, Synthesis, 2002, 1053.
23. X. G. Guo and M. D. Watson, Org. Lett., 2008, 10, 5333.
24. G. G. Malliaras, J. R. Salem, P. J. Brock and C. Scott, Phys. Rev. B: Condens. Matter, 1998, 58, 13411.
25. C. Goh, R. J. Kline, M. D. McGehee, E. N. Kadnikova and J. M. J. Fréchet, Appl. Phys. Lett., 2005, 86.
26. B. T. de Villers, C. J. Tassone, S. H. Tolbert and B. J. Schwartz, J. Phys. Chem. C, 2009, 113, 18978.
27. M. Pomerantz, Tetrahedron Lett., 2003, 44, 1563.
28. M. Pomerantz and A. S. Amarasekara, Synth. Met., 2003, 135–136, 257.
29. H. Meier, U. Stalmach and H. Kolshorn, Acta Polym., 1997, 48, 379.
30. M. Köse, Theoretical Chemistry Accounts: Theory, Computation, and Modeling Theoretica Chimica Acta., 2011, 128, 157.
31. J. Pommerehne, H. Vestweber, W. Guss, R. F. Mahrt, H. Bassler, M. Porsch and J. Daub, Adv. Mater., 1995, 7, 551.
32. M. Thelakkat and H. W. Schmidt, Adv. Mater., 1998, 10, 219.
33. M. Helgesen and F. C. Krebs, Macromolecules, 2010, 43, 1253.
34. F. Dierschke, A. C. Grimsdale and K. Müllen, Macromol. Chem. Phys., 2004, 205, 1147.
35. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery,Jr.,T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian, Inc., Wallingford CT, 2004.
36. J. S. Moon, C. J. Takacs, C. Shinuk, R. C. Coffin, H. Kim, G. C. Bazan and A. J. Heeger, NanoLett., 2010, 10, 4005–4008.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c1ra00570g/

---