Photovoltaic poly(rod-coil) polymers based on benzodithiophene-centred A–D–A type conjugated segments and dicarboxylate-linked alkyl non-conjugated segments

Abstract

Different from the well-studied photovoltaic conjugated polymers and small molecular compounds, poly(rod-coil) polymers are emerging as a new class of photovoltaic materials. Since they are composed of definite conjugated and non-conjugated segments in an alternative fashion, this kind of material is expected to merge the merits from both small molecular compounds and conjugated polymers. Based on benzodithiophene-centered acceptor–donor–acceptor (A–D–A) conjugated segments and dicarboxylate-linked alkyl non-conjugated segments, this study has newly designed and synthesized two poly(rod-coil) polymers. Together with three previously reported analogues, these polymers have been systematically investigated for their photovoltaic performances, with special attention paid to the effect of the dicarboxylate linking unit in non-conjugated segments and the alkyl side chains on rigid conjugated segments. It was found that the former factor has a small influence, while the latter has a significant impact on most film-related properties of the material, including film absorption spectrum, frontier orbital energy levels, bandgap, microstructure and morphology of pristine and photovoltaic blend films, as well as hole mobility. After optimization, bulk heterojunction organic solar cells based on this series of polymers reported power conversion efficiencies in range of 0.4-1.09%.

---

Introduction

Among various next-generation solar power producing technologies, solution-processable organic solar cells (OSCs) have attracted considerable attention due to their unique advantages in terms of light weight, mechanical flexibility, low cost, and adaptability to roll-to-roll large area fabrication.^1–4 With two decades' effort since the invention of the bulk heterojunction (BHJ) device configuration in 1995,⁵ the power conversion efficiency (PCE) of OSCs has been progressively raised from ∼3% (ref. 6) to nowadays over 10% (ref. 7–11) for a single junction cell, making this technology more realistic for its real use. Looking back this amazing OSC history, one may recognize that one of important driving forces in the field comes from persistent material innovation, especially in high performance donor materials.

In the early era, conjugated homopolymers represented by poly(phenylenevinylene)s and poly(thiophene)s, were mainly engaged as donor materials and achieved a PCE level of 3-5%.^12–15 Later, donor material innovation shifted to conjugated copolymers composed of alternative electron-donating (D) and electron-accepting (A) units.^16–18 Owing to their special backbone structures, D–A conjugated copolymers generally have an intense intramolecular charge-transfer (ICT) absorption band in visible and even near infrared region, thus enable a more efficient solar light harvesting. Moreover, numerous D and A units available and their various combination ways allow producing a variety of materials with finely tuned energy levels for both highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), interchain interactions, and charge mobility. Consequently, a number of D–A conjugated copolymers with high performances have been invented, and thus substantially improved PCE record over 10%.^7–9 However, researchers found that the so-developed conjugated polymer photovoltaic materials, including both homo- and D–A types, often suffered a severe molecular weight-dependent photovoltaic behavior. For example, Brabec, Schilinsky, and their coworkers found that poly(3-hexylthiophene)-based solar cells showed a PCE exceeding 2.5% only when its number-average molecular weight (M_n) is larger than 10 000.¹⁹ When its molecular weight decreased to several thousands, the device PCE dramatically dropped and was lower than 0.5%. In another work reported by Bazan et al., poly[(4,4-didoceyldithieno[3,2-b:2′3′-d]silole)-2,6-diyl-alt-(2,1,3-benzothiadiazole)-4,7-diyl], a typical high performance D–A conjugated copolymers, displayed a PCE of 5.9% at M_n of 34 kDa, but only 1.2% at M_n of 7 kDa.²⁰ Such high molecular weight-dependent photovoltaic performance of conjugated polymers would cause a severe batch-reproducibility problem for their real use since this polymer parameter always varies from batch to batch.

Unlike polymers that are mixtures of homologues, small molecular compounds have a definite chemical structure and can be purified straightforwardly. Therefore, one of main material innovation activities in the present field is focused on the development of small molecular photovoltaic materials.^21-23 Following the knowledge from D–A conjugated polymers, a variety of small molecular compounds containing both D and A units have been designed and studied. Some of them showed high photovoltaic performances with a record PCE value larger than 10%.¹⁰ However, since they usually have a large rigid π-conjugated core, these kinds of compounds are prone to aggregate or crystallize in solid state, and hard to form a well-qualified flat film, especially in a large size.

In 2014, we proposed a new kind of photovoltaic materials named poly(rod-coil) polymers, which backbones are composed of definite conjugated segments and non-conjugated segments in an alternative fashion.²⁴ In such polymer chains, only conjugated segments are photo-active and would mainly contribute basic optoelectronic properties for the material. Furthermore, chemical structures of either conjugated or non-conjugated segments in all polymer chains are completely same irrespective of their molecular weight. Therefore, this kind of materials is expected to display both merits of small molecular compounds and conjugated polymers, i.e., molecular weight-insensitive optoelectronic properties and good film formation potential (originating from their polymeric nature). In our previous two works, we have demonstrated that these expectations are reasonable.^24,25 In the first work, we observed two poly(rod-coil) polymers having almost one-fold molecular weight difference displayed similar optoelectronic properties and photovoltaic performance.²⁴ In the second work, we proved that poly(rod-coil) polymers have a superior film formation capability than small molecular reference compound having a core structure identical to their rigid segments.²⁵ Soon after our first publication, this material concept has also been reported in field-effect transistors (OFETs).^26–28 More recently, Sivula et al. found that a poly(rod-coil) polymer based on 3,6-bis(5-(benzofuran-2-yl)thiophen-2-yl)-2,5-bis(2-ethylhexyl)pyrrolo[3,4-c]pyrrole-1,4(2H,5H)-dione conjugated segments and flexible alkylene non-conjugated segments can act as additive to improve device thermal stability for both OFETs and OSCs based on its reference small molecular compound.²⁹ Although the present reported performances were still very low, the emergence of such poly(rod-coil) polymeric materials opens a new avenue for developing novel optoelectronic materials having both merits of small molecular compounds and polymers. However, as compared with conventional conjugated polymers and small molecular compounds, this kind of materials have more complicated structure parameters, including all the factors on both conjugated and non-conjugated segments. Therefore, more effort is required on their design and optimization toward high performance.

In this article, we report optoelectronic properties and photovoltaic performance of a new series of poly(rod-coil) polymers, as shown in Fig. 1. Here, the skeleton of rigid conjugated segments takes an A–D–A structure using benzodithiophene (BDT) and benzothiadiazole (BT) as D and A units, respectively. Since both BDT and BT are popular building blocks and have been engaged in construction of numerous conjugated polymers and small molecular materials with high performances,^30–32 the so-designed conjugated segment is expected to endow this series of poly(rod-coil) polymers a good photovoltaic performance. In fact, we have demonstrated that a small reference compound having identical skeleton structure but with R₁ = H, R₂ = 2-ethylhexyl, and hexyl at both ends have a good photovoltaic performance with a PCE of 4.53%.³³

Fig. 1 Chemical structures of poly(rod-coil) polymers.

For non-conjugated segments, dicarboxylate-linked two 1,6-hexylene chains were adopted for the purpose of easy synthesis. Here, the dicarboxylate group varied from p-phenylenedicarboxylate, m-phenylenedicarboxylate, to 1,6-hexylenedicarboxylate, affording polymer P1, P2, and P3, respectively. To be honest, these polymers were already reported in our previous paper.²⁵ However, the theme of that paper was mainly focusing on material film formation properties and did not report any photovoltaic data. In order to investigate the effect of this dicarboxylate structural parameter on material properties and photovoltaic performance, these three polymers were included in this work again and some of their data were cited again for comparison.

Besides the structural parameter of non-conjugated segments, this work also paid special attention on the alkyl side chains of rigid conjugated segments. It is well known that side chain engineering is vitally important to achieve high performance conjugated polymer and small molecular photovoltaic materials.^34–37 In general, soluble side chains not only help the material getting better solution processability, but also strongly affect film microstructures, including interchain or intermolecular interactions, chain/molecular packing structures and their preferential orientation. Thus, as compared with P2, polymer P4 and P5 having different R₁ and R₂ on rigid conjugated segments were newly designed and synthesized. Detail photovoltaic investigation and careful comparisons among these five polymers revealed that alkyl side chains on rigid conjugated segments have profound impact on material property and performance, but dicarboxylate linking group has little influence.

Results and discussion

The synthesis of P4 and P5 followed the same method for preparation of P1–P3.²⁵ As shown in Scheme 1, thiophene was firstly lithiated with BuLi and then reacted with tetrahydropyranyl (THP)-protected 1-bromohexyl alcohol, affording compound 1 in a yield of 66%. Then, compound 1 was subjected to stannylation and subsequently one equivalent Stille coupling with 4,7-dibromo[2,1,3]benzothiadiazole to produce compound 3, one important intermediate in the synthesis. Based on this intermediate, compound 6 was prepared by attaching one thiophene unit via Suzuki coupling reaction and deprotection of THP with p-toluenesulfonic acid (PTSA). After Mitsunobu esterification with isophthalic acid mediated by diisopropyl azodicaroxylate (DIAD) in the presence of PPh₃, the important dibromo monomer, compound 7, was obtained in a yield of 78%. The final polymers P4 and P5 were synthesized via Stille coupling polymerization of dibromo monomer 7 with trimethylstannylated BDT monomers having different alkoxy side chains in toluene at 110 °C in yield of 75% and 68%, respectively. Their number-average molecular weight (M_n) and polydispersity (PDI) analysed by gel permeation chromatography (GPC) are listed in Table 1. The data show that molecular weights of both P4 and P5 are in the range of several thousands, comparable to those of P1–P3. All these polymers except P4 have good solubility in common chlorinated solvents, such as chloroform (CF), chlorobenzene (CB) and o-dichlorobenzene (ODCB) at room temperature. However, polymer P4 is only soluble in hot solvents, because it has no alkyl side chains at R₁ positions and smaller alkyl substituents at R₂ positions (2-ethylhexyl vs. 2-hexyldecyl in polymer P5). Thermogravimetric analysis (TGA) reported thermal decomposition temperature (T_d) in nitrogen atmosphere is 327 °C for polymer P4 and 324 °C for polymer P5 (Table 1, Fig. S9†). Although both values are slightly lower than those of P1–P3, they are larger than 300 °C, indicating polymer P4 and P5 are stable enough for optoelectronic device applications.

Scheme 1 Synthesis of P4 and P5.

Table 1 Basic properties of poly (rod-coil) polymer P1–P5

Polymer	Mn^a (kDa)	PDI^a	Td^b (°C)	λmax (nm)		Eg,opt (eV)	Eox,onset (V)	Energy level (eV)
				Solution	Film			HOMO	LUMO
a Determined by GPC using polystyrene standards and THF eluent.b 5%-Weight loss temperature.c Data from ref. 25.
P1^c	6.6	1.6	342	320, 400, 532	309, 404, 566	1.67	0.52	−5.13	−3.46
P2^c	6.7	1.9	336	319, 400, 532	303, 397, 564	1.68	0.51	−5.12	−3.44
P3^c	8.4	2.0	336	320, 400, 532	309, 404, 566	1.68	0.50	−5.11	−3.43
P4	3.7	1.2	327	402, 533	332, 415, 557	1.75	0.66	−5.27	−3.52
P5	5.2	1.5	324	315, 401, 532	314, 405, 545	1.87	0.61	−5.22	−3.35

---

The UV-vis absorption spectra of these poly(rod-coil) polymers are shown in Fig. 2 and their data are summarized in Table 1. In dilute chlorobenzene solution, the five polymers all showed two absorption bands from 350 to 650 nm with two peak-tops at around 400 and 532 nm. The former band can be assigned to π-π* transition of rigid conjugated segments, while the latter one originates from ICT transition between electron-accepting BT unit and electron-donating BDT unit. Since these solution spectra are almost identical for all studied polymers, it is indicated that both dicarboxylate linking unit in non-conjugated segments and side chains at R₁ and R₂ position of conjugated segments have little influence on material light absorption properties in solution.

Fig. 2 UV-vis absorption spectra of polymer P1–P5 (a) in chlorobenzene solution and (b) in film state.

In film state, all the polymers displayed red-shifted absorption spectra, as compared with those in solution (Fig. 2b). Of interest, the red-shift amount of ICT band was found to highly depend on polymer structure, which was 32-34 nm for P1, P2, and P3, while decreased to 24 nm for P4 and then to 13 nm for P5 (Table 1). Furthermore, the film absorption spectra of P1, P2, and P3, kept the same shape as those in solution. In contrast, spectral shape changed in the cases of P4 and P5, in which a clear absorption shoulder appeared around 600 nm in film state. The almost same ICT redshift values and keeping same spectral shape in film state observed in the cases of P1, P2, and P3 indicate that different dicarboxylate linking units in non-conjugated segments may not interfere interactions and packing behaviours among conjugated segments. However, the alkyl side chains on rigid conjugated segments have significant influence. The appearance of absorption shoulder around 600 nm for both P4 and P5 may imply the removal of hexyl chains at R₁ position changes the packing structure among rigid conjugated segments. While, the replacement of R₂ substituents with larger alkyl chains may tend to decrease π–π interactions among conjugated segments, as reflected by smaller ICT redshift observed for P5 than P4. Similarly, such side chains also affect optical bandgap (E_g,opt) of the material, that is, 1.75 eV for P4 and 1.87 eV for P5, both larger than those of P1–P3 (around 1.68 eV).

In order to get energy level information for both HOMO and LUMO of the polymers, cyclic voltammetry (CV) was performed with their film samples in a three-electrode-configuration cell using a glass carbon as working electrode, a Pt wire as counter electrode and AgNO₃/Ag as reference electrode. As shown in Fig. 3 and Table 1, all polymer samples displayed an irreversible oxidation peak in positive scanning region. Their onset potentials (E_ox,onset) were detected to be 0.50-0.52 V for P1, P2, and P3, while 0.66 V for P4 and 0.61 V for P5 as reference to Ag⁺/Ag electrode. Under the same conditions, ferrocenium/ferrocene (Fc⁺/Fc) redox couple appeared at 0.19 V. Based on the standard energy level of Fc⁺/Fc couple (−4.8 eV) and the above measured E_ox,onset data, polymer HOMO energy levels were calculated to be −5.11–5.13 eV for P1, P2 and P3, while −5.27 eV for P4 and −5.22 eV for P5. Again, the data indicate that different dicarboxylate linking units in non-conjugated segments do not alter redox potential of the material, and thus has ignorable influence on HOMO energy level. In contrast, the removal of hexyl side chains in R₁ position of rigid conjugated segments positively shifts the material oxidation potential, therefore affording a deeper HOMO energy level. Moreover, the size of R₂ substituents also showed some influence on onset oxidation potential and HOMO energy level. Since OSC open-circuit voltage (V_OC) is semiempirically in proportion to energy level difference between HOMO of donor material and LUMO of acceptor material, the deeper HOMO energy level for P4 and P5 would be favour for their photovoltaic applications.

Fig. 3 Cyclic voltammograms of P1–P5.

The film structure was studied by means of X-ray diffraction (XRD) technology on film samples. As shown in Fig. 4, P1 and P2 films did not display any obvious diffraction peak except two broad halos centred at 5.42 and 24.1° with d-spacings of 16.28 and 3.69 Å, indicating the absence of regular chain packing structure in these films. In contrast, P3 film gave a strong diffraction peak at 3.48° (d-spacing: 25.34 Å) and two similar broad halos at 5.41 and 24.2°. While in the case of P5 film, these two halo peaks became sharp and strong with clear peak-tops at 5.09 and 23.4°. The former peak generally accounted for the presence of a kind of layered interchain packing structure with an interchain distance of 17.34 Å. The latter has a d-spacing of 3.80 Å, which can be assigned for (010) diffraction peak and suggests the existence of strong π–π interactions in the film. All these results indicate both dicarboxylate units in non-conjugated segments and alkyl side chains of rigid conjugated segments affect chain packing structure in film state. The removal of hexyl chains at R₁ positions enhances π–π interactions among rigid conjugated segments. However, in differential scanning calorimetry (DSC) measurements, no thermal transition was observed for all the studied polymers (Fig. S10†).

Fig. 4 XRD profiles of drop-casted neat films of P1, P2, P3, and P5.

Photovoltaic performance was investigated using conventional BHJ cells with an architecture of ITO/PEDOT:PSS/active layer/Ca/Al. The active layer was composed by a blend film of the checked polymer and [6,6]-phenyl-C₆₁-butyric acid methyl ester (PC₆₁BM). In order to find best performance of all the studied polymers except P4, device fabrication conditions, including polymer/PC₆₁BM weight ratio, solvent, spin-coating speed, annealing process, were firstly optimized (Table S1–S3 and S5†). It was found the best polymer/PC₆₁BM weight ratio is to be 1 : 1.5 for P1, P2, P3 and P5. For processing solvent, ODCB is suitable for P1–P3, while chloroform would be first choice for P5. Owing to the poor solubility of P4, hot ODCB was chosen as processing solvent for fabrication of P4-based cells and device optimization was only performed on spin-coating speed (Table S4†). Fig. 5a displays current density-voltage (J–V) curves of all the best devices for each polymer and their devices parameters are summarized in Table 2. The data show that the solar cells based on P1, P2, and P3 displayed a V_OC in the range of 0.73-0.76 V, a J_SC in the range of 1.99-2.19 mA cm⁻², and a FF value in the range of 28.8-33.8%, thus giving a PCE in the range of 0.46-0.51%. The so small difference in these device parameters suggests dicarboxylate linking units in non-conjugated segments have no significant effect on material photovoltaic performance. In comparison, the devices based on P4 and P5 showed larger values in both V_OC and J_SC, affording a larger PCE than those based on P1, P2 and P3. This indicates that the removal of hexyl chains at R₁ positions of rigid conjugated segments substantially improve material photovoltaic performance. Furthermore, one can find that FF value enhanced from 31.6% for P4-based cell to 49.4% for P5-based cell. This improvement finally leaded to P5 cell achieved a PCE of 1.09%, the best value among all checked devices.

Fig. 5 (a) J–V curves and (b) EQE spectra of the best OSCs based on P1–P5 in blending with PC₆₁BM.

Table 2 Device parameters of OSCs as shown in Fig. 5a and hole mobilities of their active films

Polymer	VOC (V)	JSC^a (mA cm^−2)	FF (%)	PCE^b (%)	μh (10^−5 cm^2 V^−1 s^−1)
a Data in parentheses were calculated from EQE spectra shown in Fig. 5b.b Data in parentheses are average PCE values.
P1	0.73	2.19 (2.13)	29.8	0.47 (0.44)	0.81
P2	0.76	1.99 (1.89)	33.8	0.51 (0.50)	1.21
P3	0.74	2.15 (2.06)	28.8	0.46 (0.42)	0.52
P4	0.79	2.77 (2.80)	31.6	0.70 (0.64)	4.31
P5	0.83	2.67 (2.54)	49.4	1.09 (1.00)	3.84

---

In order to get insight into photovoltaic performances of these polymers, hole mobility was measured firstly using a space-charge limited current (SCLC) method. The measurements were performed with hole-only devices having a structure of ITO/PEDOT:PSS/active layer/Au, in which active layers have the same composition as those for OSC devices. As shown in Fig. S11† and Table 2, the active layer based on P4 and P5 in blending with PC₆₁BM showed a hole mobility of 4.31 × 10⁻⁵ and 3.84 × 10⁻⁵ cm² V⁻¹ s⁻¹, respectively. Both values are larger than those for the active layers based on P1 (0.81 × 10⁻⁵ cm² V⁻¹ s⁻¹), P2 (1.21 × 10⁻⁵ cm² V⁻¹ s⁻¹), and P3 (0.52 × 10⁻⁵ cm² V⁻¹ s⁻¹). This fact indicates the removal of hexyl chains at R₁ positions of rigid conjugated segments can improve hole transportation. Such improvement implies that P4 and P5 films have stronger π–π interactions among conjugated segments and more regular packing structure than those of P1–P3, coinciding with aforementioned observations in film absorption spectroscopy and XRD measurements.

Secondly, the active layer morphology was investigated by atomic force microscopy (AFM) in a tapping mode. As shown in Fig. 6a and c, the topographies of P1/PC₆₁BM and P3/PC₆₁BM based active layers looked like the piling of certain particle-like objects. In their phase images (Fig. 6f and h), many black spots obviously appeared over an earth yellow background, which also seemed to be composed by interconnected particles. Such observations suggest that both components in these active layers aggregated into particle-like objects. In the case of P2/PC₆₁BM active layer, its topographic image displayed a picture full of worm-like objects (Fig. 6b). In its phase image (Fig. 6g), many short black rods (formed by interconnected black spots) in addition to isolated black spots were observed over an earth yellow background, which also changed to have a continuous phase configuration. These views imply that both P2 and PC₆₁BM components have an incline to form continuous phases in this blend film. However, in the case of P4/PC₆₁BM film, many large particle-like objects appeared again, as revealed by its topographic and phase images (Fig. 6d and i). Since their topographic appearances and phase shifts were different from the above black spots, such large particles are deduced to be P4 polymer aggregates owing to its poor solubility. Finally, the situation has been changed in the case of P5/PC₆₁BM active layer. As shown by topographic and phase images in Fig. 6e and j, this active layer displayed a discontinuous biphasic separation structure. Although domain widths for both phases were found in the range of several tens to one hundred nanometer and larger than ideal 20 nm, such biphasic separation structure undoubtedly was the best morphology among all the checked active layers, and thus endowed P5/PC₆₁BM-based cell the best photovoltaic performance. All these results clearly demonstrate both dicarboxylate linking units in non-conjugated segments and alkyl side chains in rigid conjugated segment have significant impact on the morphology and microstructure of the active layer.

Fig. 6 Tapping-mode AFM (a–e) topographic and (f-j) phase images of the active layers based on (a and f) P1, (b and g) P2, (c and h) P3, (d and i) P4, and (e and f) P5 in blending with PC₆₁BM.

Conclusions

In this work, two poly(rod-coil) polymers bearing BDT-centred A–D–A rigid conjugated segments and dicarboxylate-linked alkyl chains as non-conjugated segments have been designed and synthesized. These two polymers together with the previous reported three analogues have been investigated in detail on their photovoltaic properties with special attention on the effect of dicarboxylate-linking units in non-conjugated segments and alkyl chains on rigid conjugated segments. It was found that different dicarboxylate linking units in non-conjugated segments have little influence on material optical properties, HOMO and LUMO energy levels, and material bandgap. Although they somewhat affected pristine film structure and blend film morphology, their solar cells did not exhibit big different performance. In contrast, alkyl chains on conjugated segments have been demonstrated to have significant impact on the above film-related properties. The removal of hexyl chains at R₁ positions and the replacement of 2-ethylhexyl with 2-hexyldecyl chains at R₂ positions afforded the rod-coil polymer that showed the best photovoltaic performance among the series. Compared with other polymers, such polymer exhibited relatively stronger π–π interchain or inter-segment interactions, more ordered pristine film structure, a larger hole mobility, and a better blend film morphology. As the best device efficiency was only 1.09%, the photovoltaic performance of poly(rod-coil) polymers are still lag far behind conjugated polymers and small molecular materials. Much effort and further exploration are needed for achieving this kind of photovoltaic materials with high performances.

Experimental section

Materials

Unless indicated, all commercial reagents were used as received. All reactions were carried out in an Ar atmosphere with freshly distilled dry solvents, which were dehydrated following standard methods. Compound 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran,³⁸ 2,6-bis(trimethyltin)-4,8-di(2-ethylhexyloxy)benzo[1,2-b;3,4-b′]di-thiophene,³⁹ and 2,6-bis(trimethyltin)-4,8-di(2-hexyl)decyloxy-benzo[1,2-b;3,4-b′]dithiophene,³⁹ were synthesized following the reported methods.

2-((6-(Thiophen-2-yl)hexyl)oxy)tetrahydro-2H-pyran (compound 1). To a THF solution (50 mL) of thiophene (2.45 g, 30.9 mmol), n-BuLi (1.6 M in n-hexane, 14.5 mL, 23.2 mmol) was added dropwise at −78 °C. After stirred at −78 °C for 1 h, the reaction mixture was added with 2-((6-bromohexyl)oxy)tetrahydro-2H-pyran (4.87 g, 18.5 mmol). Afterward, the reaction mixture was warmed to room temperature naturally and stirred at room temperature overnight. After quenched with water, the mixture was extracted with dichloromethane for 3 times. The combined organic layers were then dried with anhydrous Na₂SO₄, filtered and concentrated by rotary evaporation in a reduced pressure. The residue was subjected to silica gel column chromatography using n-hexane/CH₂Cl₂ (4/1, v/v) as eluent to get compound 1 as colourless oil. Yield: 3.28 g (66%). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 7.10 (d, J = 4.0 Hz, 1H), 6.91 (t, J = 4.0 Hz, 1H), 6.77 (d, J = 3.6 Hz, 1H), 4.56 (t, J = 4.0 Hz, 1H), 3.84-3.89 (m, 1H), 3.70-3.76 (m, 1H), 3.47-3.52 (m, 1H), 3.35-3.41 (m, 1H), 2.82 (t, J = 8.0 Hz, 2H), 1.50-1.73 (m, 10H), 1.36-1.44 (m, 4H).

Tributyl(5-(6-((tetrahydro-2H-pyran-2-yl)oxy)hexyl)thiophen-2-yl)stannane (compound 2). Compound 1 (3.05 g, 11.4 mmol) was dissolved in 46 mL dry THF under the protection of argon. The solution was cooled down to −78 °C using a dry ice-acetone bath, and n-butyllithium (1.6 M in n-hexane, 9.25 mL, 14.8 mmol) was added dropwise. The solution was slowly warmed up to room temperature and stirred for 2 h. Afterwards, the mixture was cooled down to −78 °C again, and tributyltin chloride (1 M in n-hexane, 17.1 mL, 17.1 mmol) was added in one portion, and the reaction mixture was allowed to slowly warm to room temperature and stirred overnight. Then, 20 mL of cold water was poured into the flask, and solution was extracted with diethyl ether. Organic layers were combined, washed with brine, dried over anhydrous MgSO₄, and concentrated under vacuum. The obtained crude product (6.36 g, colorless oil) was used for the next step without further purification.¹H NMR (400 MHz, CDCl₃) δ (ppm): 6.97 (d, J = 4.0 Hz, 1H), 6.89 (d, J = 4.0 Hz, 1H), 4.57 (t, J = 4.0 Hz, 1H), 3.84-3.89 (m, 1H), 3.70–3.76 (m, 1H), 3.47–3.51 (m, 1H), 3.35-3.41 (m, 1H), 2.86 (t, J = 8.0 Hz, 2H), 1.50-1.73 (m, 23H), 1.01-1.12 (m, 5H), 0.84-0.96 (m, 13H).

4-Bromo-7-(5-(6-((tetrahydro-2H-pyran-2-yl)oxy)hexyl) thiophene-2-yl)benzo[c][1,2,5]thiadiazole (compound 3). To a two-necked flask was added 4,7-dibromobenzo[c][1,2,5]thiadiazole (3.36 g, 11.4 mmol), compound 2 (6.35 g, 11.37 mmol), Pd(PPh₃)₄ (138.7 mg, 0.12 mmol), and 60 mL dry toluene. After degassing by three freeze–pump–thaw cycles and filling back with Ar, the reaction mixture was heated to reflux for 24 h. Afterwards, cold water was added to quench the reaction, followed by extraction with chloroform/water for several times. All organic layers were collected, dried over anhydrous Na₂SO₄, filtered, and concentrated by rotary evaporation in a reduced pressure. The residue was subjected to silica gel column using CHCl₃ as eluent to get compound 3 as dark red solid. Yield: 3.42 g (62% for two steps). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 7.92 (d, J = 4.0 Hz, 1H), 7.82 (d, J = 8.0 Hz, 1H), 7.63 (d, J = 8.0 Hz, 1H), 6.87 (d, J = 4.0 Hz, 1H), 4.57 (t, J = 4.0 Hz, 1H), 3.86 (m, 1H), 3.73 (m, 1H), 3.51 (m, 1H), 3.39 (m, 1H), 2.88 (t, J = 6.4 Hz, 2H), 1.4–1.9 (m, 14H).

4-(5-(6-((Tetrahydro-2H-pyran-2-yl)oxy)hexyl)thiophen-2-yl)-7-(thiophen-2-yl)benzo[c][1,2,5]thiadiazole (compound 4). A mixture of thiophen-2-yl boronic acid (2.43 g, 19.0 mmol), compound 3 (3.05 g, 6.33 mmol) and Pd(PPh₃)₄ (69.3 mg, 0.06 mmol) was dissolved in 60 mL THF. After K₂CO₃ aqueous solution (3 M, 20 mL) was added, the reaction mixture was subjected to degas by freeze-pump-thaw cycles, filled back with Ar, and refluxed for 20 h. After the reaction mixture was extracted with chloroform/water, the organic layers were collected, dried over Na₂SO₄, filtered, and concentrated by rotary evaporation in a reduced pressure. The residue was subjected to silica gel column chromatography using n-hexane/CH₂Cl₂ (1/1, v/v) as eluent, allowing to get compound 4 as red solid. Yield: 2.93 g (92%). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 8.08 (d, J = 4.0 Hz, 1H), 7.92 (d, J = 4.0 Hz, 1H), 7.82 (d, J = 8.4 Hz, 1H), 7.76 (d, J = 10.0 Hz, 1H), 7.42 (d, J = 6.8 Hz, 1H), 7.18 (m, 1H), 6.85 (d, J = 4.0 Hz, 1H), 4.55 (t, J = 4.0 Hz, 1H), 3.85 (m, 1H), 3.72 (m, 1H), 3.48 (m, 1H), 3.38 (m, 1H), 2.87 (t, J = 8.0 Hz, 2H), 1.4-1.8 (m, 14H).

6-(5-(7-(Thiophen-2-yl)benzo[c][1,2,5]thiadiazol-4-yl)-thiophene-2-yl)hexan-1-ol (compound 5). To a CHCl₃ solution (35 mL) of compound 4 (1.43 g, 2.95 mmol) was added a MeOH solution (5.0 mL) of p-toluene sulfonic acid (PTSA, 516.9 mg, 2.95 mmol) dropwise. After stirred at 50 °C for 2 h, the reaction mixture was washed with water for several times, dried over Na₂SO₄, filtered, and concentrated by rotary evaporation in a reduced pressure. The residue was subjected to silica gel column chromatography using CH₂Cl₂ as eluent, allowing to get compound 5 as red solid. Yield: 1.03 g (87%). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 8.08 (d, J = 4.4 Hz, 1H), 7.93 (d, J = 4.4 Hz, 1H), 7.83 (d, J = 10.0 Hz, 1H), 7.76 (d, J = 10.0 Hz, 1H), 7.44 (d, J = 6.8 Hz, 1H), 7.19 (t, J = 5.6 Hz, 1H), 6.86 (d, J = 4.0 Hz, 1H), 3.65 (t, J = 8.4 Hz, 2H), 2.89 (t, J = 10.0 Hz, 2H), 1.74-1.79 (m, 2H), 1.28-1.61 (m, 8H).

6-(5-(7-(5-Bromothiophen-2-yl)benzo[c][1,2,5]thiadiazol-4-yl)-thiophen-2-yl)hexan-1-ol (compound 6). To a THF solution (120 mL) of compound 5 (1.0 g, 2.5 mmol) was added N-bromosuccinimide (NBS, 578.4 mg, 3.25 mmol) in one portion at 0 °C under dark. After stirred at room temperature overnight, the reaction mixture was concentrated by rotary evaporation in a reduced pressure. The residue was subjected to a silica gel column chromatography using CH₂Cl₂ as eluent to get the crude product, which was further purified by recrystallization from ethanol. Finally, 1.07 g pure compound 6 was obtained as red solid. Yield: 89%. ¹H NMR (400 MHz, CDCl₃) δ (ppm): 7.95 (d, J = 3.6 Hz, 1H), 7.78 (d, J = 4.0 Hz, 3H), 7.14 (d, J = 4.4 Hz, 1H), 6.88 (d, J = 4.0 Hz, 1H), 3.63-3.69 (m, 2H), 2.89 (t, J = 4.0 Hz, 2H), 1.74-1.81 (m, 2H), 1.58-1.61 (m, 2H), 1.41-1.46 (m, 4H).

Bis(6-(5-(7-(5-bromothiophen-2-yl)benzo[c][1,2,5]thiadiazol-4-yl)thiophen-2-yl)hexyl) isophthalate (compound 7). To a THF solution (30.0 mL) of isophthalic acid (79.3 mg, 0.48 mmol), compound 6 (479.9 mg, 1.00 mmol), and triphenylphosphine (3.13 g, 11.9 mmol) was added diisopropyl azodicarboxylate (DIAD, 2.2 mL, 11.9 mmol) slowly at 0 °C. After stirred at 0 °C for 1 h, then at room temperature overnight, the reaction mixture was concentrated by rotary evaporation in a reduced pressure. The residue was subjected to a silica gel column chromatography using n-hexane/CH₂Cl₂ (1/2, v/v) as eluent to get compound 7 as red solid. Yield: 404.3 mg (78%). ¹H NMR (400 MHz, CDCl₃) δ (ppm): 8.67 (s, 1H), 8.20-8.24 (m, 2H), 7.91 (d, J = 3.6 Hz, 2H), 7.72-7.74 (m, 6H), 7.52 (t, J = 7.6 Hz, 1H), 7.12 (d, J = 4.0 Hz, 2H), 6.85 (d, J = 3.6 Hz, 2H), 4.35 (t, J = 6.4 Hz, 4H), 2.88 (t, J = 4.0 Hz, 4H), 1.78-1.80 (m, 4H), 1.48-1.53 (m, 8H), 1.21-1.30 (m, 4H). LRMS (MALDI) m/z: 1088.4 (M⁺).

P4. To a two-necked flask was added dibromo-monomer compound 7 (91.2 mg, 0.084 mmol), 2,6-bis(trimethyltin)-4,8-di(2-ethylhexyloxy)benzo[1,2-b;3,4-b′]dithiophene (64.8 mg, 0.084 mmol), Pd(PPh₃)₄ (1.23 mg, 0.001 mmol) and 10 mL dry toluene. After degased by three freeze–pump–thaw cycles and filled back with Ar, the reaction mixture was heated to reflux for 48 h. After cooling to room temperature, the reaction mixture was poured into 100 mL methanol to precipitate the crude polymer. After filtration, the crude product was obtained and further subjected to Soxhlet extraction with methanol, acetone, hexane, and chloroform in sequence. The chloroform extraction fraction was collected, evaporated to dryness, and further dried in vacuum for 1 day to get the final polymer product as red solid. Yield: 120 mg (75%).

P5. P5 was synthesized as red solid in a yield of 68% following the same method of P4 but using 2,6-bis(trimethyltin)-4,8-di(2-hexyl)decyloxy-benzo[1,2-b;3,4-b′]dithiophene in the place of 2,6-bis(trimethyltin)-4,8-di(2-ethyl)hexyloxy-benzo[1,2-b;3,4-b′]dithio-phene.

Characterizations and measurements

¹H NMR spectra were recorded on a Varian Mercury 400 MHz spectrometer using CDCl₃ as a solvent and tetramethylsilane (TMS) as an internal reference. Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectroscopy was carried out on a Shimadzu Biotech Axima Performance Mass Spectrometer using dithranol or α-cyano-4-hydroxycinnamic acid as a matrix. Gel permeation chromatography (GPC) was carried out on a Waters 1515 HPLC instrument equipped with a Waters 2489 UV detector, using THF as an eluent. The molecular weight and polydispersity index (PDI) were calculated based on polystyrene standards. UV-vis absorption spectroscopy was performed on a Hitachi U-3310 spectrophotometer. Cyclic voltammetry (CV) measurements were performed on a CHI 660C instrument using a three-electrode cell with a glassy carbon as working electrode, a platinum wire as counter electrode, and Ag/AgNO₃ as reference electrode. The samples were first casted on a glassy carbon electrode to form a film and then measured in CH₃CN in the presence of 0.1 M Bu₄NPF₆ with a scan rate of 50 mV s⁻¹. Thermogravimetric analysis (TGA) was carried out by a TGA Q500 instrument under N₂ with a temperature rate of 10 °C min⁻¹. Differential scanning calorimetry (DSC) was performed on a Q2000 modulated DSC instrument under N₂ with a heating rate of 10 °C min⁻¹ and a cooling rate of 15 °C min⁻¹. X-ray diffraction (XRD) was carried out on a PANalytical X'Pert Pro diffractometer with Cu K_α beam (40 kV, 40 mA) in θ–2θ scans (0.033 Å step size, 30 s per step). Polymer sample films were prepared by drop-casting from their CHCl₃ solutions onto a quartz plate.

Fabrication and characterization of OSC devices

OSC devices in the work adopted an architecture of ITO/PEDOT:PSS/active layer/Ca/Al. Firstly, ITO glass substrates were cleaned sequentially in ultrasonic baths with detergent, deionized water, acetone and isopropyl alcohol for 20 min, dried over nitrogen gas and cleaned with a UV-O₃ cleaner for 20 min. Then, a layer of poly(3,4-ethylenedioxy-thiophene):poly(4-styrenesulfonate) (PEDOT:PSS, Heraeus Clevios PVP. Al 4083) was spin-coated onto the cleaned ITO-coated substrate at 4000 rpm and baked at 140 °C for 15 min in air. Afterwards, the active layer was fabricated onto PEDOT:PSS layer by spin-coating a solution of checked polymer and PC₆₁BM in desired solvent and concentration and subjected to certain annealing treatment. Finally, the device was accomplished by subsequently thermal deposition of Ca/Al electrode (10/100 nm) on the top of the active layer at 2 × 10⁻⁶ Torr through a shadow mask. The so-prepared devices had an effective area of 7 mm². Layer thickness was measured on a Veeco Dektak 150 profilometer. Current density–voltage (J–V) curves were recorded with a Keithley 2420 source meter. Photocurrent was acquired upon irradiation using an AAA solar simulator (Oriel 94043A, 450 W) with an AM 1.5G filter. The intensity was adjusted to be 100 mW cm⁻² under the calibration with a NREL-certified standard silicon cell (Orial reference cell 91150). External quantum efficiency (EQE) was detected with a 150 W Xe lamp, an Oriel monochromator 74125, an optical chopper, a lock-in amplifier and a NREL-calibrated crystalline silicon cell.

Hole mobility measurements

Hole mobility was measured by space-charge-limited current (SCLC) method with a device configuration of ITO/PEDOT:PSS/active layer/Au. All checked devices were fabricated following the same method as that for OSC devices but thermal deposition of a layer of Au electrode in place of Ca/Al electrode. The compositions of all checked active layers were the same as those for OCS devices. According to Mott–Gurney law, SCLC theory can be described by
where J is current density, ε_o is permittivity of vacuum, ε_r is relative permittivity of the material (for polymer, 3 in general), μ is mobility, V_a is applied voltage, V_bi is built-in voltage, and d is the thickness of the active film.

Acknowledgements

This work was supported by International Science and Technology Cooperation Program of China (No. 2015DFG62680), Science and Technology Commission of Shanghai Municipality (No. 13JC1407000), National Natural Science Foundation of China (No. 21074147), Chinese Academy of Sciences, and Zhongzhou University.

References

1. C. J. Brabec, M. Heeney, I. McCulloch and J. Nelson, Chem. Soc. Rev., 2011, 40, 1185.
2. X. Zhao and X. Zhan, Chem. Soc. Rev., 2011, 40, 3728.
3. L. Dou, Y. Liu, Z. Hong, G. Li and Y. Yang, Chem. Rev., 2015, 115, 12633.
4. L. Lu, T. Zheng, Q. Wu, A. M. Schneider, D. Zhao and L. Yu, Chem. Rev., 2015, 115, 12666.
5. G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science, 1995, 270, 1789.
6. F. Padinger, R. Rittberger and N. Sariciftci, Adv. Funct. Mater., 2003, 13, 85.
7. J.-D. Chen, C. Cui, Y.-Q. Li, L. Zhou, Q.-D. Ou, C. Li, Y. Li and J. X. Tang, Adv. Mater., 2015, 27, 1035.
8. Y. Liu, J. Zhao, Z. Li, C. Mu, W. Ma, H. Hu, K. Jiang, H. Lin, H. Ade and H. Yan, Nat. Commun., 2014, 5, 5293.
9. V. Vohra, K. Kawashima, T. Kakara, T. Koganezawa, I. Osaka, K. Takimiya and H. Murata, Nat. Photonics, 2015, 9, 403.
10. B. Kan, M. Li, Q. Zhang, F. Liu, X. Wan, Y. Wang, W. Ni, G. Long, X. Yang, H. Feng, Y. Zuo, M. Zhang, F. Huang, Y. Cao, T. P. Russell and Y. Chen, J. Am. Chem. Soc., 2015, 137, 3886.
11. J. Huang, J. H. Carpenter, C.-Z. Li, J.-S. Yu, H. Ade and A. K.-Y. Jen, Adv. Mater., 2016, 28, 967.
12. K. Tajima, Y. Suzuki and K. Hashimoto, J. Phys. Chem. C, 2008, 112, 8507.
13. Y. Kim, S. Cook, S. Tuladhar, S. Choulis, J. Nelson, J. R. Durrant, D. Bradley, M. Giles, I. Mcculloch, C. Ha and M. Ree, Nat. Mater., 2006, 5, 197.
14. W. Ma, C. Yang, X. Gong, K. Lee and A. J. Heeger, Adv. Funct. Mater., 2005, 15, 1617.
15. S. Berson, B. R. De, S. Bailly and S. Guillerez, Adv. Funct. Mater., 2007, 17, 1377.
16. M. H. Chen, J. Hou, Z. Hong, G. Yang, S. Sista, L. M. Chen and Y. Yang, Adv. Mater., 2009, 21, 4238.
17. N. Leclerc, A. Michaud, K. Sirois, J. F. Morin and M. Leclerc, Adv. Funct. Mater., 2006, 16, 1694.
18. Y. Liang, Z. Xu, J. Xia, S. T. Tsai, Y. Wu, G. Li, C. Ray and L. Yu, Adv. Mater., 2010, 22, E135.
19. P. Schilinsky, U. Asawapirom, U. Scherf, M. Biele and C. J. Brabec, Chem. Mater., 2005, 17, 2175.
20. R. C. Coffin, J. Peet, J. Rogers and G. C. Bazan, Nat. Chem., 2009, 1, 657.
21. B. Walker, A. Tamayo, X. Dang, P. Zalar, J. Seo, A. Garcia, M. Tantiwiwat and T.-Q. Nguyen, Adv. Funct. Mater., 2009, 19, 3063.
22. Y. Sun, G. Welch, W. Leong, C. Takacs, G. Bazan and A. J. Heeger, Nat. Mater., 2012, 11, 44.
23. J. Zhou, X. Wan, Y. Liu, Y. Zuo, Z. Li, G. He, G. Long, W. Ni, C. Li, X. Su and Y. Chen, J. Am. Chem. Soc., 2012, 134, 16345.
24. H.-J. Li, J.-T. Wang, C.-Y. Mei and W.-S. Li, Chem. Commun., 2014, 50, 7720.
25. W. Shao, L. Liang, X. Xiang, H.-J. Li, F.-G. Zhao and W.-S. Li, Chin. J. Chem., 2015, 33, 847.
26. A. Gasperini, S. Bivaud and K. Sivula, Chem. Sci., 2014, 5, 4922.
27. Y. Zhao, X. Zhao, M. Roder, G. Qu, Y. Diao, A. L. Ayzner and J. Mei, Chem. Mater., 2015, 27, 7164.
28. Y. Zhao, Xi. Zhao, Y. Zang, C.-A. Di, Y. Diao and J. Mei, Macromolecules, 2015, 48, 2048.
29. A. Gasperini, X. A. Jeanbourquin, A. Rahmanudin, X. Yu and K. Sivula, Adv. Mater., 2015, 27, 5541.
30. Z. C. He, C. M. Zhong, S. J. Su, M. Xu, H. B. Wu and Y. Cao, Nat. Photonics, 2012, 6, 591.
31. J. Zhou, Y. Zuo, X. Wan, G. Long, Q. Zhang, W. Ni, Y. Liu, Z. Li, G. He, C. Li, B. Kan, M. Li and Y. Chen, J. Am. Chem. Soc., 2013, 135, 8484.
32. C.-Y. Chang, Y.-J. Cheng, S.-H. Hung, J.-S. Wu, W.-S. Kao, C.-H. Lee and C.-S. Hsu, Adv. Mater., 2012, 24, 549.
33. L. Liang, J.-T. Wang, X. Xiang, J. Ling, F.-G. Zhao and W.-S. Li, J. Mater. Chem. A, 2014, 2, 15396.
34. L. Yang, H. Zhou and W. You, J. Phys. Chem. C, 2010, 114, 16793.
35. C. Cabanetos, A. E. Labban, J. A. Bartelt, J. D. Douglas, W. R. Mateker, J. M. Frechét, M. D. McGehee and P. M. Beaujuge, J. Am. Chem. Soc., 2013, 135, 4656.
36. I. Osaka, M. Saito, T. Koganezawa and K. Takimiya, Adv. Mater., 2014, 26, 331.
37. I. Osaka, Y. Houchin, M. Yamashita, T. Kakara, N. Takemura, T. Koganezawa and K. Takimiya, Macromolecules, 2014, 47, 3502.
38. M. Voigt, M. Klaumünzer, A. Ebel, F. Werner, G. Yang, R. Marczak, E. Spiecker, D. M. Guldi, A. Hirsch and W. Peukert, J. Phys. Chem. C, 2011, 115, 5561.
39. C.-Y. Mei, L. Liang, F.-G. Zhao, J.-T. Wang, L.-F. Yu, Y.-X. Li and W.-S. Li, Macromolecules, 2013, 46, 7920.

---

Footnotes

† Electronic supplementary information (ESI) available: ¹H NMR spectra of compound 1-7, MALDI-TOF-MS of compound 7, TGA, DSC, SCLC, and device performance of all fabricated solar cells. See DOI: 10.1039/c6ra01200k

‡ These authors contributed equally.

---