Simple synthesis of novel terthiophene-based D–A₁–D–A₂ polymers for polymer solar cells

Abstract

Direct arylation was used to synthesize a series of novel terthiophene (T3)-based D–A₁–D–A₂ polymers and D–A₂–D monomers in fewer synthetic steps. In these T3-based D–A₁–D–A₂ polymers, pyrrolo[3,4-c]pyrrole-1,4-dione (DPP) was selected as the first acceptor A₁, octyl-thieno[3,4-c]pyrrole-4,6-dione (TPD) or 2,1,3-benzothiadiazole (BT) or fluorinated benzothiadiazole (FBT) was selected as the secondary acceptor A₂. T2-based polymer with the bithiophene segments (T2) as the donor was synthesized for comparison, too. UV-vis absorption, electrochemical properties, blend film morphology, and photovoltaic properties of the polymers were studied to explore the effects of the oligothiophene unit and secondary acceptor moiety (A₂), meanwhile, the fluorine substitution effect was also discussed. It is shown that the change of donor segment from T2 to T3 introduces a difference in the energy levels, crystallinity, polymer:PC₇₁BM morphology and PSC performances between the T2-based and T3-based D–A₁–D–A₂ polymers. Varying the secondary acceptor (A₂) from BT to TPD also promotes the crystallinity and backbone planarity leading to enhanced PSC performances of the T3-based D–A₁–D–A₂ polymer. Although the effectiveness of fluorine substitution for tuning the UV-vis absorption, energy levels and degree of crystallinity has been demonstrated, the insufficient E^DONOR_LUMO − E^PCBM_LUMO energy offset and poor miscibility of polymer:PC₇₁BM limit the short circuit current (J_sc). In addition, the highest J_sc of 12.98 mA cm⁻² is achieved for P1, while the higher HOMO level limits the open circuit voltage (V_oc) and leads to a power conversion efficiency (PCE) of 4.36%.

---

Introduction

Bulk heterojunction (BHJ)-type solar cells have been extensively studied and power conversion efficiencies (PCEs) of 9–12% have been reported.^1–3 These achievements were mainly attributed to the advance in electron-rich unit (D)–electron-deficient unit (A) alternating low band gap polymers.^4,5 To date, a number of successful electron-deficient units have been identified, such as 2,1,3-benzothiadiazole (BT), fluorinated benzothiadiazole (FBT), pyrrolo[3,4-c]pyrrole-1,4-dione (DPP) and alkylthieno[3,4-c]pyrrole-4,6-dione (TPD).^6–8 However, developing new D and A units has been more difficult and the normal D–A copolymers synthesized present two absorption bands, which are not very broad.

Recently, regioregular terpolymers have emerged for an alternative molecular design strategy of conjugated polymers with distinct optoelectronic properties.^9–11 These terpolymers comprise three components with two different donor units and one acceptor unit or with one donor unit and two different acceptor units on their conjugated polymer backbones. Therefore, terpolymers allow the integration of the advantages of different donor and acceptor unit to obtain well-controlled physical properties. Thus, a new class of D–A₁–D–A₂ polymers was developed. Several research works demonstrate that the D–A₁–D–A₂ polymer design strategy is very promising for producing highly efficient polymers for solar cells.^12–14 Moreover, it is shown that alternating D–A₁–D–A₂ polymer having regioregular structure forms π-stacking structure and hence provides higher carrier mobility compared with a random copolymer with two acceptor units.

The streamlined synthesis of conjugated polymers is an important research area in which one of the major goals is the development of simpler, faster, and hence cheaper procedures.¹⁵ Stille cross-coupling is a popular method often used for the polycondensation of thiophene-based monomers, but highly toxic organo-tin halides are commonly used for the synthesis of monomers, which themselves are toxic as well. Suzuki coupling is well established and environmental, but entails mostly the use of aryl bromides and boron-based monomers, which require either cryogenic temperatures or catalytic conditions. Direct arylation (DA) is an emerging technique that potentially overcomes all of these drawbacks. This method has obvious advantages, such as fewer synthetic steps, less time, higher monomer purity, less harmful toxic substance.^16,17 Compared to other coupling reaction, DA can create new aromatic C–C linkages from acyl halides and activated aromatic C–H bonds. Many reports have detailed the use of DA method for the synthesis of conjugated polymers with a large variety of small molecule building blocks.^18–20 However, few reports have studied the use of DA method for producing small molecule and polymer based D–A₁–D–A₂ materials for PSCs application. The comparative study between the DA and Suzuki coupling method of preparing the regular D–A₁–D–A₂ polymer for PSCs was reported in our previous work.²¹ The regular D–A₁–D–A₂ polymers via incorporating two electron-deficient moieties (DPP and BT) with bithiophene (T2) as the electron-donating units by using the DA were synthesized and our work demonstrated that the efficient D–A₁–D–A₂ polymer for solar cells can be simply synthesized by optimized DA polymerization conditions.

Recently, an effective method which is to tailor the length of the donor segment along the main chain for varying the energy band of D–A polymer has been reported. Some works reported that extending the length of the oligothiophene donor segment can manipulate the energy band and terthiophene-based polymers with higher PCE were obtained.^22–24 In addition, fluorine substitution or variation the electron-acceptor moieties could tune the optoelectronic properties of the D–A₁–D–A₂ polymer.

Based mentioned above, we designed simple synthesis of terthiophene (T3)-based D–A₁–D–A₂ polymers by DA, in which terthiophene segments were between the two electron-withdrawing moieties. The design strategy is shown in Fig. 1. Here, D–A₁–D and D–A₂–D intermediates were synthesized as the building blocks. In these building blocks, DPP was selected as the first acceptor A₁, TPD or BT or FBT was selected as the secondary acceptor A₂. Two thiophene units were inserted into DPP for DPP building block (2T–DPP–2T) by Stille coupling. Hexyl-thiophene was inserted into the acceptor (A₂) for the second building block because electron-donating hexyl side chain located at the 3 and 3′ position can promote the DA polymerization and affect the solubility and molecular packing of the polymer. These D–A₂–D monomers (BrHT–TPD–HTBr, BrHT–BT–HTBr, BrHT–FBT–HTBr) had been prepared as reported and used as the promising acceptor moieties for the high performance polymers with D–π–A structures.^25–27 However, they are mostly synthesized with the costly borylation or stannylation. Here, direct arylation of 2-bromo-3-hexylthiophene with dibromo-TPD (BT or FBT) unit is carried out to give the monomer in one step. Then, three T3-based D–A₁–D–A₂ polymers are synthesized with D–A₁–D and D–A₂–D monomers by DA in fewer synthetic steps. For comparison, the T2-based polymer with the bithiophene segments (T2) as the donor was synthesized. UV-vis absorption, electrochemical properties, film morphology, and photovoltaic properties of the polymers were studied to explore the effects of the oligothiophene unit and secondary acceptor moiety (A₂), meanwhile, fluorine substitution effect was also discussed.

Fig. 1 The design strategy of T3-based D–A₁–D–A₂ polymers.

Results and discussion

Synthesis and thermal property

Four intermediates 3,6-di([2,2′-bithiophen]-5-yl)-2,5-bis(2-octyldodecyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione (2T–DPP–2T), 3-bis(5-bromo-4-hexylthiophen-2-yl)-5-octyl-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (BrHT–TPD–HTBr), 4,7-bis(5-bromo-4-hexylthiophen-2-yl)benzo[c][1,2,5]thiadiazole (BrHT–BT–HTBr), 4,7-bis(5-bromo-4-hexylthiophen-2-yl)-5,6-difluorobenzo[c][1,2,5]thiadiazole (BrHT–FBT–HTBr), were synthesized (ESI†). The monomer T–DPP–T as starting materials, the bromination and Stille coupling afforded the monomer 2T–DPP–2T. The yield from T–DPP–T to monomer 2T–DPP–2T is acceptable (∼70%), and the purification is convenient. 2-Bromo-3-hexylthiophene, which undergoes optimized direct arylation reaction to obtain monomer BrHT–TPD–HTBr, BrHT–BT–HTBr and BrHT–FBT–HTBr with a low yield (25–30%) because the additional oligomerization takes place. Although the yield is low, the monomer synthesis is faster, cheaper and nontoxic compared with existing routes.

The T3-based D–A₁–D–A₂ polymers were prepared via DA polymerization in which DPP (A₁)-based monomer (2T–DPP–2T) was as C–H monomer and the second acceptor (A₂)-based monomer (BrHT–TPD–HTBr or BrHT–BT–HTBr or BrHT–FBT–HTBr) was as C–Br monomers. In comparison, T2-based polymer was prepared with T–DPP–T as C–H monomer and BrHT–TPD–HTBr as C–Br monomer. It has been shown that steric hindrance can have a profound influence on the effectiveness of C–C cross coupling reactions.²⁸ Compared to C–H monomer T–DPP–T, monomer 2T–DPP–2T as C–H monomer has more thiophene moieties resulting in negative steric influence during DA polymerization. In order to promote the DA based 2T–DPP–2T, the selectivity of C–Br monomer plays an important role. Thus, the A₂-based monomers with the hexyl chain at thiophene moiety were used as C–Br comonomer. Here, electron-donating hexyl side chain located at the 3 and 3′ position can increase the reactivity and selectivity of C–Br monomer and promote the DA polymerization.²⁹ meanwhile, it can obviously affect the solubility and molecular packing of the copolymer. Under the DA condition (K₂CO₃) (0.25 mmol), pivalic acid (0.1 mmol), catalyst (Pd(OAc)₂) (5 mol%), ligand (tricyclohexylphosphonium tetrafluoroborate) (10 mol%), the target T3-based D–A₁–D–A₂ polymers were successfully obtained. All polymers were purified by Soxhlet extraction by using methanol, acetone, n-hexane to remove the oligomers. Then the polymers were collected by using chloroform as the extraction to discard the insoluble solids. Note that polymer P1 and P4 were selected to compare the differences in the properties between T2-based and T3-based polymers, because they have the same side chains and type of acceptor and allow a fair comparison. Compared to T2-based polymer P4, T3-based D–A₁–D–A₂ polymers had lower relative molecular weight values. It is very likely due to the expected higher steric hindrance. However, M_n value greater than 12 kg mol⁻¹ were achieved for the each T3-based polymer, with M_w values ranging from ∼21.0 kg mol⁻¹ to 35 kg mol⁻¹, which enabled them as the donor material of active layer.

The thermal properties of the polymers were investigated using thermal gravimetric analysis (TGA) and were carried out under nitrogen atmosphere at a heating rate of 10 °C min⁻¹. As shown in Fig. S15,† the onset decomposition temperature (T_d) (5% weight loss) was found to be at 384 °C (P1), 392 °C (P2), 366 °C (P3) and 385 °C (P4), respectively. All of the polymers have good thermal stability. Good thermal stability retards the deformation of the polymer morphology and the degradation of the active layer at elevated temperatures, which are desirable qualities for application as photoactive layers for PSCs.³⁰

Optical and electrochemical properties

The normalized UV-vis absorption spectra of the four polymers in CHCl₃ solutions and thin films are shown in Fig. 2. The absorption maxima wavelength (λ_max), the optical band gap deduced from the solution and film absorption onsets are summarized in Table 1. All these ternary D–A₁–D–A₂ polymers showed relatively broad absorption (350–1000 nm) and full width at half-maximum (>240 nm) in thin films, where the high energy absorbance band located at the range of 350–500 nm was originated from the π–π* transition of polymer backbones, while the low energy absorbance band located at the range of 500–1000 nm with strong intensity could be attributed to the intra-molecular charge transfer (ICT) from the donor unit to the acceptor unit in the polymer backbone.³¹ Compared to the solutions, the absorption peaks of the corresponding films of the polymers (P2, P3, P4) were broadened and red-shifted. Noted that λ_max of polymer P1 afforded a blue-shift from solution to film due to the formation of molecular packing and solvent effect in the P1 solution. Compared with that of T2-based polymer P4 film, the low-energy absorption band (500–1000 nm) of P1 slightly blue-shifted. When the oligothiophene segments extended from T2 to T3, the effective density of acceptor unit was decreased, and the electronic effect of the acceptor unit was thus decreased, which led to higher HOMO and LUMO levels and also blue-shifted absorption.²³ The shoulder peaks at the longer wavelength in film were obviously observed in P4 and P1, which demonstrated that π–π stacking interaction of ordered molecular packing existed in the solid state. Compared with that of P2, the absorption range of P1 obviously broadened, meanwhile, the maximum absorption peak largely red-shifted. This can be probably explained by the fact that replacing BT unit with TPD unit as the A₂ unit is beneficial for high degree of planarity of the polymer backbone leading to an extended effective π-conjugation length. The broader absorption and smaller bandgaps should help to achieve larger short circuit current (J_sc). The fluorinated polymer P3 displayed red-shifted and broader absorption compared to non-fluorinated polymer P2. This is expected, since the addition of fluorine substituent, in most cases, lowers the HOMO and LUMO levels, resulting in red shift in the absorption profile. The shoulder peaks appeared in P2 and P3, too, which indicated these two T3-based polymers existed ordered molecular packing. From the onset of film absorption, the optical bandgaps of P1–P4 are estimated to be 1.23 eV, 1.28, 1.21, 1.22 eV, respectively.

Fig. 2 UV-vis spectra of P1, P2, P3 and P4 in chloroform solution (a) and film (b).

Table 1 The UV-vis spectra data and energy levels of four polymers

Polymer	Solution^a (nm)		Film^b (nm)			E^optg^d eV	Ered eV	Eoxd eV	HOMO^e eV	LUMO eV	E^cvg^f eV
	λmax	λonset	λmax	Fwhm^c	λonset
a Polymer in dilute CHCl3 solution (c = 1.0 × 10^−5 mol L^−1).b Thin film processed with CHCl3 solution.c Full width at half-maximum in film.d Estimated from the absorption band edge in film, E^optg = 1240/λonset (eV).e EHOMO = −e(Eoxd + 4.42) V; ELUMO = −e(Ered + 4.42) V.f E^cvg = ELUMO − EHOMO.
P1	763	981	740	275	1011	1.23	−0.86	0.70	−5.12	−3.56	1.56
P2	677	860	711	240	972	1.28	−0.84	0.72	−5.14	−3.58	1.56
P3	693	880	729	253	1021	1.21	−0.67	0.80	−5.22	−3.75	1.47
P4	691	890	751	260	1019	1.22	−0.78	0.82	−5.24	−3.64	1.60

---

The electrochemical properties of the polymer thin films were characterized by cyclic voltammetry (CV). The highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) energy levels were determined from their onset oxidation (E_oxd) and reduction potentials (E_red) according to the empirical relationship proposed by de Leeuw et al.³² E_HOMO = −e(E_oxd − E_Fc/Fc⁺ + 4.80) V, E_LUMO = −e(E_red − E_Fc/Fc⁺ + 4.80) V. By assuming the energy level of ferrocene/ferrocenium (Fc/Fc⁺) to be 4.8 eV below the vacuum level,³³ the formal potential of Fc/Fc⁺ was measured as 0.38 V against Ag/Ag⁺.³⁴

The CV curves of the polymers are shown in Fig. 3. And the respective HOMO and LUMO energy levels are summarized in Table 1. Compared to P4, P1 displayed higher HOMO and LUMO levels, the reason was discussed in above. As the A₂ unit changed from TPD unit to BT unit, the energy levels were similar. Meanwhile, the HOMO and LUMO energy levels of fluorinated polymer P3 were lower than that of non-fluorinated polymer P2 due to the electron-withdrawing nature of the fluorine atoms. However, the LUMO level of the fluorinated polymer P3 was −3.75 eV, which might be too deep to provide a sufficient LUMO offset with PC₇₁BM (LUMO, −4.0 eV), because it is commonly believed that a LUMO offset of 0.3 eV is needed to ensure highly efficient exciton dissociations for polymer:PCBM cells.

Fig. 3 Cyclic voltammetric curves and energy levels of P1, P2, P3 and P4.

Photovoltaic properties

The photovoltaic properties of the polymers were investigated in BHJ solar cell with a device architecture of ITO/PEDOT:PSS/polymers:PC₇₁BM/PFN/Al. PC₇₁BM was used as the electron acceptor material due to its stronger absorption in the visible region (440–530 nm) than PC₆₁BM; this choice could compensate for the absorption valley of polymers.³⁵ In order to obtain the maximum value of PCE, the performance of solar cell was carefully optimized for each polymer by varying the solvent, the polymer:PC₇₁BM weight ratio, the polymer concentration, the spin coating speed and the volume of additive (1,8-diiodooctane DIO). A thin layer of PFN was used as the cathode buffer layer, which could effectively enhance the electron collection of the PSCs.^36,37 The current density–voltage (J–V) and the external quantum efficiency (EQE) of the PSCs are shown in Fig. 4(a) and (b), and the corresponding photovoltaic data from the J–V curves are listed in Table 2.

Fig. 4 J–V characteristics (a) and EQE spectra (b) of PSCs based P1, P2, P3 and P4.

Table 2 The device data with polymers

Polymer	Voc (V)	Jsc (mA cm^−2)	FF (%)	PCE (%)	EQEmax (%)	E^polymerLUMO − E^PCBMLUMO (eV)
P1	0.58	12.98	58.3	4.36	0.51	0.44
P2	0.61	9.33	64.5	3.66	0.43	0.42
P3	0.62	1.60	45.1	0.45	0.12	0.25
P4	0.72	5.93	65.3	2.79	0.34	0.36

---

As shown in Fig. 4(a) and Table 2, the PCE of three T3-based polymers varied with the different secondary acceptor (A₂). For P1 and P2, high J_sc values (9–13 mA cm⁻²) were obtained, as was characteristic of low bandgap polymers and DPP-based materials in general. The highest PCE of the device based T3-based polymer P1 was found to be 4.36%, with a V_oc of 0.58 V, a J_sc of 12.98 mA cm⁻² and a FF of 58.27%. Compared to T2-based polymer P4, the devices based T3-based P1 had higher PCE owing to the high J_sc. The relatively low V_oc of P1 can be explained by its higher HOMO levels. On the other hand, changing A₂ unit of TPD to BT decrease the J_sc of the devices based P2. The good charge transport ability and morphology of active layer could contribute to the improved J_sc of the device based P1. Compared to T2-based polymer with similar molecular weight in our previous work,²¹ P2 displayed enhanced PCE (3.05% → 3.66%). These results indicated that extending donor segment from T2 to T3 could enhance device performance. However, J_sc value obtained for P3 was significantly lower, 1.60 mA cm⁻², resulting in a low PCE. This is contradictory to the fact that the mostly fluorinated polymers have been shown to lead to improved OPV devices.³⁸ Even so, a few works have reported that fluorination of conjugated polymers could complicate the morphology of the active layer and the electronic properties and did not guarantee better performance.¹⁹ Moreover, we noticed that P3 displayed broader absorption relative to the non-fluorinated polymer P2, the reduction in J_sc might be related to the E^DONOR_LUMO − E^PCBM_LUMO energy level offset. For P3, the E_LUMO offsets with PC₇₁BM were 0.25 eV, which was significantly below the 0.3 V offset threshold design criteria for most materials designed to function with PC₇₁BM. P3 exhibited relatively low V_oc although it showed a deep HOMO level, which illustrated that there were a number of factors besides HOMO levels that could influence V_oc. In addition, the bad morphology of active layer is likely factor in the lower observed J_sc in P3 and P4 as will be discussed later.

To investigate the origin of the J_sc, the corresponding EQE spectra of all the devices were recorded under illumination by monochromatic light. As shown in Fig. 4(b), it was realized that all integrals of the EQEs were in good accordance with the J_sc values that had been measured from corresponding devices for all of the resultant polymers. Compared to the device based T2-based polymer P4, devices based T3-based polymer (P1 and P2) exhibited broader and stronger response range coving from 300–900 nm, which meant that the overall photocurrent were mainly contributed by the PC₇₁BM absorption (300–600 nm) and the polymer absorption (500–900 nm). Devices based P1 demonstrated the most efficient photo-response with a higher EQE in a broad range of 350–850 nm that benefited from the intensive absorption in this region. This result agrees with the highest J_sc of the corresponding device based P1. In contrast, the devices based fluorinated-polymer P3 showed negligible contribution in the polymer absorption range and a small response for PC₇₁BM leading to the lowest J_sc.

Molecular stacking and active layer morphology

To provide further insight into the morphology of the polymers, X-ray diffraction measurement (XRD) was performed on drop-cast films onto a glass substrate to study the effect of structural changes on the polymer stacking. Fig. 5 showed the XRD patterns of polymers. The diffraction peaks at small and large 2θ zone could be ascribed to the laminar packing and π–π stacking between polymer backbones, respectively. The sharp diffraction peaks of P1–P4 were at 4.04°, 4.46°, 4.12°, 3.83°, corresponding to inter-chain d-spacing of 2.19 nm, 1.97 nm, 2.15, 2.30 nm, respectively. P1 and P3 exhibited higher intensity with higher order diffraction peaks, indicating a higher degree of crystallinity. The weak diffraction peaks were at 23.30°, 22.79°, 23.00°, 22.60°, corresponding to π–π stacking spacing of 3.81 Å, 3.91 Å, 3.87 Å, 3.93 Å, respectively. The results indicate that FBT or TPD as the secondary acceptor in these T3-based D–A₁–D–A₂ polymers can promote backbone planarity and thus, a closer π–π stacking. This is in accordance with the UV-absorption discussion. In general, small π–π spacing is preferable since it reduces the energy barrier for interchain charge hopping.³⁹

Fig. 5 XRD curves of P1, P2, P3 and P4.

Tapping mode atomic force microscopy (AFM) gave us insight into the surface morphology of ternary D–A₁–D–A₂ polymer:PC₇₁BM blend films (Fig. 6 and S16†). The surface roughness (R_RMS) of the AFM topographic images were 3.48 nm (P1:PC₇₁BM), 4.55 nm (P2:PC₇₁BM), 5.58 nm (P3:PC₇₁BM), 4.68 nm (P4:PC₇₁BM), respectively. As shown in Fig. 6 and S16,† relatively uniform and smooth morphology with less aggregations and the proper and continuous phase separation were observed in P1 (P2):PC₇₁BM film which was essentially favorable for the charge carrier transport and could potentially reduce the probability of charge recombination within the photoactive layer. This was in accordance with the higher J_sc obtained in the devices based P1 and P2. The distinct clusters and rougher morphology were observed on the surface of the P3:P₇₁CBM film, which might originate from the poor miscibility of the polymer P3 with PC₇₁BM. The large domain phase separation often leads to greater charge recombination and inefficient charge separation. It is clear that in addition to the insufficient E^DONOR_LUMO − E^PCBM_LUMO energy offset caused by fluorination, the coarse and large phase separation of P3 blend can also limit the photocurrent. In contrast, P4 blend film exhibited relatively larger aggregation, which might be the reason for the lower J_sc of the device of P4 (5.93 cm mA⁻²).

Fig. 6 AFM topographic images of polymer:PC₇₁BM blend films: P1 (a), P2 (b), P3 (c) and P4 (d).

Experimental section

Materials

All reagents and starting materials were purchased from commercial sources. 2,5-Bis(2-octyldodecyl)-3,6-di(thiophen-2-yl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione (T–DPP–T, purity: >98%), and 1,3-dibromo-5-octyl-4H-thieno[3,4-c]pyrrole-4,6(5H)-dione (5, purity: >98%) was purchased from Derthon. 4,7-Dibromo-5,6-difluorobenzo[c][1,2,5]thiadiazole (4, purity: >98%) was purchased from Suna Tech Inc. Benzo[c][1,2,5]thiadiazole (2, purity: 97%) was purchased from Sinopharm Chemical Reagent Co. Ltd. All reagents were put to use without further purification. The solvent must be corresponding treatment with distillation in order to being in the free-water and free-oxygen conditions. The synthesis details of monomers are shown in ESI.†

General procedure for synthesis of polymers

Four polymers were synthesized by palladium (2)-catalyzed direct arylation polycondensation (synthetic routes are shown in Scheme S1†). Monomer 2T–DPP–2T (0.1 mmol, 102.6 mg) or T–DPP–T (0.1 mmol, 86.1 mg), monomer BrHT–TPD–HTBr or BrHT–BT–HTBr, BrHT–FBT–HTBr (0.1 mmol), K₂CO₃ (34.6 mg, 0.25 mmol), pivalic acid (10.2 mg, 0.1 mmol), catalyst (Pd(OAc)₂) (5 mol%), ligand (tricyclohexylphosphonium tetrafluoroborate) (10 mol%) and 1 mL DMAc/p-xylene (v/v = 1 : 1) solvent were transferred to a 5 mL round-bottom flask under an atmosphere of nitrogen. The mixture was stirred for 48 h at 110 °C. After cooling to room temperature, the mixture was poured into 200 mL methanol. The precipitate was separated by filtration. The product of higher molecular weight was obtained by Soxhlet extraction from methanol, acetone and n-hexane, respectively. The residual solid was dissolved in chloroform and the solution was concentrated. The polymer was obtained by re-precipitation from methanol. The products were dried under vacuum at 40 °C overnight. Polymer P1: yield: 67.7%. ¹H NMR (400 MHz, CDCl₃) δ: 9.06 (d, 2H), 8.95 (d, 2H), 7.52 (d, 4H), 7.03 (s, 2H), 5.35 (d, 2H), 4.09 (d, 4H), 2.00–0.88 (m, 118H). GPC (THF, polystyrene standard): M_n = 12.5 kg mol⁻¹, M_w = 21.7 kg mol⁻¹, PDI = 1.74. Polymer P2: yield: 43.6%. ¹H NMR (400 MHz, CDCl₃) δ: 8.92 (d, 2H), 8.05 (d, 2H), 7.86 (d, 2H), 7.20 (s, 2H), 3.98 (d, 4H), 2.02–0.83 (m, 78H). GPC (THF, polystyrene standard): M_n = 18.0 kg mol⁻¹, M_w = 30.4 kg mol⁻¹, PDI = 1.69. Polymer P3: yield: 45.5%. ¹H NMR (400 MHz, CDCl₃) δ: 9.12 (d, 2H), 8.95 (d, 2H), 8.08 (d, 2H), 8.05 (d, 2H), 7.61 (d, 2H), 7.09 (s, 2H), 4.03 (d, 4H), 1.94–0.83 (m, 100H). GPC (THF, polystyrene standard): M_n = 24.3 kg mol⁻¹, M_w = 35.0 kg mol⁻¹, PDI = 1.44. Polymer P4: yield: 49.2%. ¹H NMR (400 MHz, CDCl₃) δ: 8.98 (d, 2H), 8.05 (d, 2H), 7.49 (d, 2H), 7.03 (s, 2H), 6.83 (d, 4H), 3.42 (m, 2H), 2.92 (d, 4H), 2.62–0.82 (m, 119H). GPC (THF, polystyrene standard): M_n = 40.5 kg mol⁻¹, M_w = 62.7 kg mol⁻¹, PDI = 1.55.

Measurement

The ¹H NMR and ¹³C NMR spectra were recorded on a BRUKER AVIII-400 NMR spectrometer by utilizing deuterated chloroform (CDCl₃) as solvent and tetramethylsilane (TMS) as standard. MALDI-TOF mass spectra were recorded on a Bruker microflex LRF spectrometer. Number-average (M_n) and weight-average (M_w) molecular weights were determined using waters 1515 gel permeation chromatography (GPC) analysis with THF as eluent and polystyrene as standard. Thermogravimetric analysis (TGA) was performed on a Mettler-Toledo 851e/822e analysis system under N₂ at a heating rate of 10 °C min⁻¹. Cyclic voltammetric (CV) measurements was carried out in a conventional three-electrode cell using a platinum plate as the working electrode, a platinum wire as the counter electrode, and Ag/Ag⁺ electrode as the reference electrode on a Zahner IM6e Electrochemical Workstation in a tetrabutylammonium hexafluorophosphate (Bu₄NPF₆) (0.1 M) acetonitrile solution at a scan rate of 20 mV s⁻¹. UV-vis absorption spectra were recorded on a SHIMADZU UV-2600 spectrometer. X-ray diffraction (XRD) data were collected using an X'Pert 3 Powder diffractometer (Philips, USA) with Cu Ka radiation. The sample films were prepared by drop-casting of polymer chloroform (CHCl₃) solution (10 mg mL⁻¹) onto glass and dried. Atomic force microscopy (AFM) images were obtained using a NanoMan VS microscope in the tapping mode. The sample films were prepared by spin-coated of polymer/PCBM solution onto ITO-coated glass and dried overnight.

Fabrication of devices and characterization

Photovoltaic devices were fabricated in a typical sandwich structure of ITO/PEDOT:PSS/polymers:PC₇₁BM/PFN/Al. ITO-coated glass substrates (15 Ω cm⁻²) were cleaned by ultrasonication sequentially detergent, distilled water, acetone, and isopropyl alcohol and then treated with O₂-plasma. After drying the substrates, PEDOT:PSS (Baytron PVP A1 4083) was spin-coated (3500 rpm for 30 s) on ITO glass and then dried at 150 °C for 15 min on a hot plate in air. Then, the substrates were transferred to a N₂-filled glove box. Polymer:PC₇₁BM dissolved in chloroform with 1,8-diiodooctane (DIO) under the optimization condition (Table S1†) and the solution was spin-coated on the ITO/PEDOT:PSS substrate. Afterwards, methanol was spin-coated on the photoactive layer for 30 s in order to remove the residual additives. Poly[(9,9-dioctyl-2,7-fluorene)-alt-(9,9-bis(3-(N,N-dimethylamino)propyl)-2,7-fluorene)] (PFN) solution (0.4 mg mL⁻¹) (cathode buffer layer) was spin-coated at 3500 rpm on the surface of active layer. The counter electrode of Al (120 nm) were successively deposited under high vacuum onto the active layer inside a nitrogen glove box. The active areas of the cells between the cathode and anode were 4 mm². Current–voltage (J–V) characteristics were recorded using a Keithley 2400 Source Meter in the dark and under 100 mW cm² simulated AM 1.5 G irradiation. The external quantum efficiency (EQE) values of the solar cells were measured using Enlitech QE-R3011 at room temperature.

Conclusions

In summary, we presented a simple synthetic route for a series of novel terthiophene (T3)-based D–A₁–D–A₂ polymers and D–A₂–D monomers using direct arylation. Compared with existing routes, the synthesis of monomers and T3-based D–A₁–D–A₂ polymers is faster, requires less synthetic steps and non-toxic. It is shown that the change of donor segment from T2 to T3 introduces difference in the energy levels, crystallinity, polymer:PC₇₁BM morphology and PSC performances between the T2-based and T3-based D–A₁–D–A₂ polymers. Varying the secondary acceptor (A₂) from BT to TPD also promotes the crystallinity and backbone planarity leading to enhance PSC performances of T3-based D–A₁–D–A₂ polymers. Although the effectiveness of fluorine substitution for tuning UV-vis absorption, energy levels and degree of crystallinity has been demonstrated, the insufficient E^DONOR_LUMO − E^PCBM_LUMO energy offset and poor miscibility of polymer:PC₇₁BM limit the J_sc. Finally, the highest J_sc of 12.98 mA cm⁻² is achieved for P1, while the higher HOMO level limits the V_oc and leads to a moderate PCE of 4.36%. The results demonstrate that changing the thiophene segment from T2 to T3 and varying the acceptor unit can be an effective strategy for the construction of D–A₁–D–A₂ polymers. Thus, this study provides opportunity for simplifying synthesis of T3-based D–A₁–D–A₂ polymer for efficient PSCs.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (21574021, 51573026), Natural Science Foundation of Fujian Province (2015J01189, 2016J01211), and the Program for Innovative Research Team in Science and Technology in Fujian Province (IRTSTFJ).

Notes and references

1. H. Zhang, H. Yao, W. Zhao, L. Ye and J. Hou, Adv. Energy Mater., 2016, 1614–6840.
2. W. Zhao, D. Qian, S. Zhang, S. Li, O. Inganäs, F. Gao and J. Hou, Adv. Mater., 2016, 28, 4734–4739.
3. Y. Lin, F. Zhao, Q. He, L. Huo, Y. Wu, T. C. Parker, W. Ma, Y. Sun, C. Wang and D. Zhu, J. Am. Chem. Soc., 2016, 138, 4955–4961.
4. J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.-C. Chen, J. Gao and G. Li, Nat. Commun., 2013, 4, 1446.
5. W. Li, K. H. Hendriks, A. Furlan, W. S. C. Roelofs, M. M. Wienk and R. A. J. Janssen, J. Am. Chem. Soc., 2013, 135, 18942–18948.
6. E. Zhou, J. Cong, K. Hashimoto and K. Tajima, Macromolecules, 2013, 46, 763–768.
7. A. Najari, S. Beaupré, P. Berrouard, Y. Zou, J. R. Pouliot, C. Lepage-Pérusse and M. Leclerc, Adv. Funct. Mater., 2011, 21, 718–728.
8. K. H. Hendriks, W. Li, M. M. Wienk and R. A. Janssen, J. Am. Chem. Soc., 2014, 136, 12130–12136.
9. W. Li, K. H. Hendriks, A. Furlan, A. Zhang, M. M. Wienk and R. A. J. Janssen, Chem. Commun., 2015, 51, 4290–4293.
10. X. Huang, G. Zhang, C. Zhou, L. Liu, Y. Jin, S. Liu, L. Ying, F. Huang and Y. Cao, Polym. Chem., 2015, 6, 4154–4161.
11. K. H. Hendriks, G. H. L. Heintges, M. M. Wienk and R. A. J. Janssen, J. Mater. Chem. A, 2014, 2, 17899–17905.
12. D. Dang, W. Chen, R. Yang, W. Zhu, W. Mammo and E. Wang, Chem. Commun., 2013, 49, 9335–9337.
13. W. Sun, Z. Ma, D. Dang, W. Zhu, M. R. Andersson, F. Zhang and E. Wang, J. Mater. Chem. A, 2013, 1, 11141–11144.
14. M. L. Keshtov, S. A. Kuklin, D. Y. Godovsky, A. R. Khokhlov, R. Kurchania, F. C. Chen, E. N. Koukaras and G. D. Sharma, J. Polym. Sci., Part A: Polym. Chem., 2016, 54, 155–168.
15. M. Leclerc and J.-F. Morin, Design and synthesis of conjugated polymers, John Wiley & Sons, 2010.
16. M. Wakioka, Y. Nakamura, Q. Wang and F. Ozawa, Organometallics, 2012, 31, 4810–4816.
17. L. G. Mercier and M. Leclerc, Acc. Chem. Res., 2013, 46, 1597–1605.
18. F. Lombeck, R. Matsidik, H. Komber and M. Sommer, Macromol. Rapid Commun., 2015, 36, 231–237.
19. P. Homyak, Y. Liu, F. Liu, T. P. Russel and E. B. Coughlin, Macromolecules, 2015, 48, 6978–6986.
20. J. R. Pouliot, L. G. Mercier, S. Caron and M. Leclerc, Macromol. Chem. Phys., 2013, 214, 453–457.
21. M. Sun, W. Wang, W. Lv, M. Lu, S. Yan, L. Liang and Q. Ling, Synth. Met., 2015, 209, 412–418.
22. Q. Tao, Y. Xia, X. Xu, S. Hedström, O. Bäcke, D. I. James, P. Persson, E. Olsson, O. Inganäs and L. Hou, Macromolecules, 2015, 48, 1009–1016.
23. H. Hu, K. Jiang, G. Yang, J. Liu, Z. Li, H. Lin, Y. Liu, J. Zhao, J. Zhang and F. Huang, J. Am. Chem. Soc., 2015, 137, 14149–14157.
24. K. H. Hendriks, G. H. L. Heintges, V. S. Gevaerts, M. M. Wienk and R. A. J. Janssen, Angew. Chem., Int. Ed., 2013, 52, 8341–8344.
25. K. Lu, J. Fang, H. Yan, X. Zhu, Y. Yi and Z. Wei, Org. Electron., 2013, 14, 2652–2661.
26. B. Wang, S.-W. Tsang, W. Zhang, Y. Tao and M. S. Wong, Chem. Commun., 2011, 47, 9471–9473.
27. N. Wang, Z. Chen, W. Wei and Z. Jiang, J. Am. Chem. Soc., 2013, 135, 17060–17068.
28. S. Schröter, C. Stock and T. Bach, Tetrahedron, 2005, 61, 2245–2267.
29. A. P. Zoombelt, J. Gilot, M. M. Wienk and R. A. J. Janssen, Chem. Mater., 2009, 21, 1663–1669.
30. Z.-G. Zhang, K.-L. Zhang, G. Liu, C.-X. Zhu, K.-G. Neoh and E.-T. Kang, Macromolecules, 2009, 42, 3104–3111.
31. M. L. Keshtov, G. D. Sharma, V. S. Kochurov and A. R. Khokhlov, Synth. Met., 2013, 166, 7–13.
32. D. M. De Leeuw, M. M. J. Simenon, A. R. Brown and R. E. F. Einerhand, Synth. Met., 1997, 87, 53–59.
33. H. Eckhardt, L. W. Shacklette, K. Y. Jen and R. L. Elsenbaumer, J. Chem. Phys., 1989, 91, 1303–1315.
34. Q. Sun, H. Wang, C. Yang and Y. Li, J. Mater. Chem., 2003, 13, 800–806.
35. Y. Yao, C. Shi, G. Li, V. Shrotriya, Q. Pei and Y. Yang, Appl. Phys. Lett., 2006, 89, 153507.
36. K. Zhang, C. Zhong, S. Liu, C. Mu, Z. Li, H. Yan, F. Huang and Y. Cao, ACS Appl. Mater. Interfaces, 2014, 6, 10429–10435.
37. Z. Kai, G. Xing, H. Fei and C. Yong, Acta Chim. Sin., 2012, 70, 2489–2495.
38. H. Bin, Z.-G. Zhang, L. Gao, S. Chen, L. Zhong, L. Xue, C. Yang and Y. Li, J. Am. Chem. Soc., 2016, 138, 4657–4664.
39. M. S. Chen, O. P. Lee, J. R. Niskala, A. T. Yiu, C. J. Tassone, K. Schmidt, P. M. Beaujuge, S. S. Onishi, M. F. Toney and A. Zettl, J. Am. Chem. Soc., 2013, 135, 19229–19236.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra18464b

---