Design and synthesis of new ultra-low band gap thiadiazoloquinoxaline-based polymers for near-infrared organic photovoltaic application

Abstract

Two D–A copolymers, F1 and F2, with fluorene and thiazole units were substituted, respectively, on a thiadiazoloquinoxaline (TDQ) unit to enhance the electron-accepting strength of TDQ. The copolymers were synthesized by a cross-coupling Stille reaction and their optical and electrochemical properties were examined, which revealed that they have ultra-low band gaps and absorption in the near-infrared. These copolymers were employed as donors along with PC₇₁BM as an electron acceptor for the fabrication of solution-processed bulk heterojunction (BHJ) polymer solar cells. After the optimization of the donor-to-acceptor weight ratio and the solvent additive (4 v% DIO as solvent additive), devices with F1:PC₇₁BM and F2:PC₇₁BM displayed power conversion efficiencies (PCEs) of 5.80% and 3.32%, respectively. Although F2 possesses a broader absorption profile compared with F1, the lower value of PCE for the F2-based device was attributed to the low LUMO offset between F2 and PC₇₁BM, which limited the exciton dissociation. The abovementioned results indicate that these copolymers can be utilized for ternary BHJ and tandem solar cells to achieve a high PCE.

---

Introduction

The development of solar cells based on polymeric materials with a bulk heterojunction (BHJ) active layer has become an impressive research area in recent years owing to their potential as an alternative source of energy.¹ The power conversion efficiency (PCE) of these solar cells has recently been increased to reach the range of 9–11% via the design of new conjugated polymers and better control of charge transport by careful adjustment of the morphology of the BHJ active layer.² It is expected that a theoretical upper limit of 15% will be approached as a result of the design and development of new absorbing materials with a reduced band gap offset, along with a high fill factor and incident photon-to-current efficiency (IPCE).³ Achieving a high PCE in polymer solar cells requires the concurrent optimization of the absorption profile, high oxidation stability and band gap of the polymer donor, as well as the alignment of its energy levels with those of the acceptor counterpart used in the BHJ active layer. A polymer with an energy band gap (1.8 > E_g > 1.10 eV) has to be designed to harvest photon energy over a broad range extending from the visible to the near-infrared region of the solar spectrum with complementary absorption to fullerene derivatives. Low-band-gap polymers possess strong absorption in both UV-visible and near-IR regions and therefore can achieve a high photocurrent in solar cells, because the PCE of PSCs is directly proportional to the photon absorption ability of the donor polymer used in the active layer.⁴ Polymers with a narrow band gap with an absorption band up to 1000 nm absorb a higher percentage of solar energy and can provide a short-circuit current density (J_sc) of over 30 mA cm⁻².⁵ The most successful strategy for designing low-band-gap polymers is the copolymerization of alternating electron-rich and electron-acceptor units for intramolecular charge transfer (ICT) between these units. ICT can be modified by adjusting the strength of the two monomers, thereby enabling tuning of the HOMO and LUMO levels.^6,7

The design and synthesis of high-performance low-band-gap polymers is quite challenging and there are few low-band-gap polymers with a PCE of higher than 8%,^4,8 because their open-circuit voltage is relatively low at about 0.7 V.⁹ The reduction of the optical band gap of a polymer is generally accompanied by either a rise in the HOMO or a reduction in the LUMO energy level. A rise in the HOMO energy level has a direct impact on the V_oc, whereas a lower LUMO energy level causes a reduction in the exciton dissociation efficiency, which is directly related to the J_sc of a photovoltaic device. Moreover, it is important to design low-band-gap polymers with both high thermal and air stability. The air stability can also be affected by the HOMO level of the polymer. Polymers with lower HOMO levels are generally more stable against oxidation in air. In addition, low-band-gap polymers should have high hole mobility in order to achieve a high PCE.¹⁰ However, all ultra-low-band-gap polymers do not directly guarantee a high PCE. Proper alignment of the HOMO and LUMO energy levels is critical for efficient charge transfer to the acceptor and to ensure a large V_oc of the device. High carrier mobility, as well as favorable morphology when blended with an electron-acceptor material, is required to enhance the device performance. Recently, a number of NIR conjugated polymers have been designed and achieved a PCE of 6% for single-junction PSCs.¹¹ These NIR narrow-band-gap polymers are also very promising for applications in tandem solar cells in combination with medium-band-gap polymers. Quite recently, it has been reported that a tandem solar cell based on an NIR conjugated polymer as one of the donor polymers reached a PCE of 10.6%.¹² Therefore, the development of NIR polymers with strong absorption in the range above 700 nm in combination with a medium band gap is very important for obtaining promising tandem solar cells with a high PCE. These recent achievements inspired us to develop novel NIR-absorbing polymers for PSC applications. Strongly electron-donating and strongly electron-accepting blocks are generally used for the development of donor–acceptor (D–A) polymers with an ultra-narrow band gap. Common strongly electron-donating monomers include pyrrole, thiophene, and bithiophene derivatives, and strongly electron-accepting blocks used in NIR D–A polymers are diketopyrrolopyrrole,¹³ benzobisthiadiazole,¹⁴ pyrazinoquinoxaline,¹⁵ thiadiazoloquinoxaline (TDQ),¹⁶ benzotriazole,¹⁷ thienoisoindigo,¹⁸ tetraazabenzodifluoranthene diimides.^19,20 Some of the more promising and less studied electron-donating monomers for the synthesis of NIR conjugated polymers are TDQ derivatives, which possess such unique properties as a high extinction coefficient, intensive absorption spectrum up to the NIR range and acceptable charge mobility. Previous TDQ-containing polymers were beneficial as active materials in organic solar cells^5,21 and organic field-effect transistors.^16a,22 However, ultra-narrow-band gap polymers based on TDQ derivatives are relatively poorly studied and at the same time are highly in demand. TDQ-containing NIR polymers with an E_g of ∼1.2 eV may extend the absorption band to the near-infrared range (over 1000 nm). Fused heteroaromatic blocks with extended π-conjugation and rigid planar structures promote the formation of π–π stacking of backbones, which exhibit enhanced absorption properties and high charge mobility. TDQ is a promising construction block for the synthesis of narrow-band-gap polymers owing to the strongly electron-accepting properties of the four imine groups in TDQ. In our efforts to develop novel NIR-absorbing D–A polymers, we started from TDQ derivatives as a central acceptor block.

We prepared two novel electron-accepting TDQ-based monomers by introducing fluorene and thiazole units into TDQ blocks to increase the electron-accepting strength of TDQ. On this basis and via a cross-coupling Stille reaction, we synthesized and characterized two new NIR D–A copolymers F1 and F2 with ultra-low band gaps (E_g = 1.24 and 1.08 eV). The new synthesized F1 and F2 copolymers possess broad (panchromatic) absorption spectra in the wavelength range of 350–1200 nm and demonstrate good solubility owing to their alkyl lateral substituents. These copolymers were employed as donors along with PCBM as electron acceptor in solution-processed BHJ PSCs. After optimization, i.e., of the donor-to-acceptor ratio and the solvent additive, F1 : PC₇₁BM (1 : 2) and F2 : PC₇₁BM (1 : 2) active layers processed with a solvent additive (SA), i.e., (4 v% DIO/CF) in solvent-based bulk heterojunction PSCs displayed PCEs of 5.80% and 3.32%, respectively. Our results show that after careful structural design such copolymers demonstrate attractive photovoltaic properties and can be used for high-performance ternary BHJ and tandem PSCs.

Experimental details

Synthesis of copolymers

The synthesis of intermediate compounds and their characterization are described in the ESI.†

8,12-Bis(5-bromo-4-dodecylthiophen-2-yl)-2,5-di(nonadec-3-yl)-[1,2,5]thiadiazolo[3,4-i]bis[1,3]thiazolo[4,5-a:5′,4′-c]phenazine (M1). NBS (99.2 mg, 0.5575 mmol) was added to a solution of 8,12-bis(4,5-didodecylthiophen-2-yl)-2,5-di(nonadec-3-yl)-[1,2,5]thiadiazolo[3,4-i]bis[1,3]thiazolo[4,5-a:5′,4′-c]phenazine (15) (339 mg, 0.2445 mmol) in THF (60 mL) and the reaction mixture was stirred at room temperature for 2 h. Then, the solvent was evaporated and the residue was purified by column chromatography on SiO₂ (hexane : CHCl₃ = 3 : 1). The target compound M1 was obtained with a yield of 295 mg (78%). ¹H NMR (400 MHz, CDCl₃, δ ppm): 8.89 (s, 2H), 3.33 (m, 2H), 2.66 (m, 4H), 2.25–1.90 (m, 8H), 1.8–1.1 (m, 102H), 0.9 (m, 12H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 176.14, 150.47, 146.39, 141.50, 137.29, 137.20, 134.89, 134.24, 130.77, 120.01, 119.95, 46.34, 35.07, 31.95, 31.93, 29.99, 29.85–29.66, 29.51, 29.40, 29.37, 28.03, 27.29, 22.70, 14.12, 11.91. Anal. calculated (%) for C₈₄H₁₃₀Br₂N₆S₅: C, 65.34; H, 8.49; N, 5.44. Found: C, 65.49; H, 8.40; N, 5.35.

4,9-Dibromo-6,7-bis(9,9-didodecyl-9H-fluoren-2-yl)-[1,2,5]thiadiazolo[3,4-g]quinoxaline (M2). Diketone 17 (1.560 g, 14.7 mmol), diamine 16 (0.573 g, 17.7 mmol) and acetic acid (50 mL) were placed into a 100 mL flask and the mixture was refluxed for 10 h. After evaporation, the residue was purified by column chromatography using a mixture of hexane : dichloromethane = 10 : 1 as eluent. The target monomer M2 was obtained as a red oil. The yield was 1.6 g (80%). ¹H NMR (400 MHz, CDCl₃, δ ppm): 7.89–7.81 (m, 4H), 7.73–7.65 (m, 4H), 7.37–7.30 (m, 6H), 1.94–1.80 (m, 8H), 1.31–0.93 (m, 72H), 0.91 (t, J = 7.0 Hz, 12H), 0.87–0.85 (m, 8H). ¹³C NMR (100 MHz, CDCl₃, δ ppm): 156.6, 152.5, 151.7, 151.0, 143.8, 140.2, 138.3, 136.6, 129.6, 128.2, 127.1, 125.2, 123.1, 120.6, 119.8, 113.8, 77.5, 77.2, 76.8, 55.3, 40.4, 32.1, 30.2, 29.8, 29.8, 29.8, 29.6, 29.5, 24.1, 22.8, 14.3. Anal. calculated (%) for C₈₂H₁₁₄Br₂N₄S: C, 73.08; H, 8.53; N, 4.16; S, 2.38; Br, 11.86. Found: C, 73.00; H, 8.44; N, 4.14; S, 2.17; Br, 11.65.

Synthesis of polymer F1. 8,12-Bis(5-bromo-4-dodecylthiophen-2-yl)-2,5-di(nonadec-3-yl)-[1,2,5]thiadiazolo[3,4-i]bis[1,3]thiazolo[4,5-a:5′,4′-c]phenazine (0.6738 g, 0.5 mmol), compound M2, 2,7-bis(trimethyltin)-4,5-diundecylbenzo[2,1-b:3,4-b′]dithiophene (0.4122 g, 0.5 mmol), Pd(Ph₃P)₄ (0.027 g, 0.0234 mmol) and toluene (20 mL) were heated under argon at 110 °C for 48 h. Then, 2-bromothiophene (0.02 g) was added and the mixture was stirred at 110 °C for another 5 h. After cooling to r.t., the reaction mixture was poured into methanol (200 mL) and filtered. The polymer was dissolved in CHCl₃ and precipitated with methanol. Then, it was purified by extraction with methanol, hexane and chloroform in a Soxhlet apparatus. The yield was 88%. Calc. for C₁₁₄H₁₆₂N₄S₃, %: C, 81.27; H, 9.69; N, 3.33; S, 5.71. Found: C, 80.96; H, 9.48; N, 2.85; S, 5.41. ¹H NMR (400 MHz, CDCl₃, δ, ppm): 9.60–7.30 (m, 16H), 3.43 (s, 8H), 2.06–0.51 (m, 138H).

Synthesis of copolymer F2. F2 was prepared similarly to F1 using 8,12-bis(5-bromo-4-dodecylthiophen-2-yl)-2,5-di(nonadec-3-yl)-[1,2,5]thiadiazolo[3,4-i]bis[1,3]thiazolo[4,5-a:5′,4′-c]phenazine (compound M1) and 2,7-bis(trimethyltin)-4,5-diundecylbenzo[2,1-b;3,4-b′]dithiophene with a yield of 79%. Calc. for C₁₁₆H₁₈₂N₆S₇, %: C, 73.91; H, 9.73; N, 4.46; S, 11.91. Found: C, 73.49; H, 9.58; N, 4.14; S, 11.54. ¹H NMR (400 MHz, CDCl₃, δ, ppm): 9.90–7.11 (m, 4H), 3.49–0.80 (m, 178H).

Device fabrication and characterization

PSCs were fabricated with a structure of ITO/PEDOT:PSS/F1 or F2:PC₇₁BM/Al using a conventional solution processing method. Indium tin oxide (ITO)-coated glass substrates were cleaned subsequently by ultrasonic treatment in detergent, deionized water, acetone, and isopropyl alcohol for 20 min each and then dried in ambient conditions. A layer of PEDOT:PSS was spin-cast (3000 rpm, 40 nm thick) onto an ITO-glass substrate and baked at 120 °C for 10 min. The active layer was spin-coated from blends with different weight ratios of F1 or F2 and PC₇₁BM in chloroform and the solvent additive (different concentrations of 1,8-diiodooctane (DIO)/chloroform (total concentration of 20 mg mL⁻¹)) at 1500 rpm for 30 s on the top of the PEDOT:PSS layer and dried for 30 min in ambient conditions. Finally, a 60 nm aluminum (Al) layer was deposited on the top of the active layer under high vacuum using a shadow mask of 20 mm². All the devices were fabricated and characterized under an ambient atmosphere without encapsulation. Hole-only and electron-only devices with ITO/PEDOT:PSS/active layer/Au and ITO/Al/active layer:PC₇₁BM/Al architectures were also fabricated in an analogous way in order to measure the hole and electron mobility, respectively. The current–voltage characteristics of the BHJ organic solar cells were measured using a computer-controlled Keithley 238 source meter under a simulated AM 1.5G spectrum at 100 mW cm⁻². A xenon light source coupled with an optical filter was used to provide the simulated irradiance at the surface of the devices. The incident photon-to-current efficiency (IPCE) of the devices was measured by illuminating the devices using the light source and a monochromator and the resulting current was measured using a Keithley electrometer under short-circuit conditions.

Results and discussion

Synthesis and characterization

The synthetic routes of the monomers M1 and M2 are illustrated in Scheme 1. To obtain M1 monomer, a new α-diketone 8 was synthesized in seven reaction steps via the following route, comprising: the synthesis of α-ethylstearic acid 2 by treatment of compound 1 with LDA, tetramethylethylenediamine (TMEDA) and ethyl iodide; the reaction of α-ethylstearic acid with carbonyldiimidazole (CDI) and an aqueous solution of ammonia with the subsequent generation of α-ethylstearic acid amide 3; followed by treatment with Lawesson's reagent with the subsequent production of α-ethylstearic acid thioamide 4. 2-(1-Ethylheptadecyl)-1,3-thiazole 5 was synthesized by the treatment of compound 4 with a bromoethyl acetal under hydrochloric acid catalysis. The alkylthiazole 5 was then converted into the corresponding bromothiazole 6 by sequential treatment with n-BuLi and CBr₄ at a reduced temperature. After ionization of derivative 6 with LDA at −78 °C, migration of the halogen to a position adjacent to the nitrogen atom occurred. The resulting organolithium intermediate underwent oxidative dimerization after the addition of an equimolar quantity of copper(II) chloride to form the dithiazole derivative 7 with a yield of 91%, which was converted into the desired 2,7-bis(1-ethylheptadecyl)-[1,3]thiazolo[4,5-g][1,3]benzothiazol-4,5-dione (8) with a yield of 37% in a one-pot reaction by sequential treatment with n-BuLi and diethyl oxalate. Furthermore, to obtain the dinitro compound 13 a Grignard reagent was prepared from dodecyl bromide and then converted into compound 11 by the introduction of an alkyl chain into 3-bromothiophene via Kumada coupling with subsequent stannylation. 4,7-Bis(4-dodecylthiophene-2-yl)-5,6-dinitrobenzo[c][1,2,5]thiadiazole 13 was prepared by treatment of compound 11 with an equimolar quantity of the dinitro derivative 12 in a Stille reaction. The diamine 14 was synthesized by the reduction of the dinitro compound 13 in acetic acid with Fe, which in reaction with the α-diketone 8 was subsequently converted into 8,12-bis(4,5-didodecylthiophene-2-yl)-2,5-di(nonadec-3-yl)-[1,2,5]thiadiazolo[3,4-i]bis[1,3]thiazolo[4,5-a:5′,4′-c]phenazine 15. The target monomer M1 was obtained by treatment of compound 15 with bromosuccinimide with a yield of 78%. Monomer M2 was synthesized from a well-known amine 16 (ref. 22b) and the α,β-diketone 1,2-bis(9,9-didodecyl-9H-fluoren-2-yl)ethane-1,2-dione (17)²³ according to Scheme 1. The reaction was conducted in acetic acid under reflux conditions. Monomer M2 was obtained after purification by column chromatography with a yield of over 80%.

Scheme 1 Synthetic route for M1 and M2 monomers.

The structures of all intermediates and final (M1 and M2) compounds were established by ¹H and ¹³C NMR spectroscopy and elemental analysis. In the aromatic part of the ¹H spectrum of compound 15, there are two singlet signals of equal intensity at δ 9.26 and 7.40 ppm, which are related to protons of the thiophene fragments. The aliphatic part of the spectrum contains signals of the alkyl substituents on the thiazole and thiophene fragments. The ratio of the integrated intensity values of the individual signals confirms the proposed structure of 15. After treatment of intermediate 15 with two equivalents of N-bromosuccinimide and the formation of the monomer M1, in the aromatic part of the spectrum only one singlet at δ 9.02 ppm was observed, which indicated that the bromination reaction was complete. The aliphatic part of the spectrum of the monomer M1 also contains signals of alkyl substituents and has integrated intensity values that are consistent with the proposed structure. The ¹³C NMR spectrum of the monomer M1 has 11 signals in the aromatic region and the most downfield signal is at δ 176.14 ppm, which corresponds to the carbon atom of the S–C–N moiety of the thiazole ring (Fig. S12, ESI†).

In the ¹H NMR spectrum of the monomer M2 in the downfield region at δ 7.89–7.81, 7.73–7.65 and 7.37–7.30 ppm there are three multiplets belonging to 14 different aromatic hydrogens; in the range of δ 1.94–1.80 ppm there are signals that are characteristic of CH₂ groups directly connected to a fluorene ring; signals belonging to the terminal CH₃ groups of alkyl chains are located at δ 0.91 ppm. Signals corresponding to the remaining hydrogen atoms of the alkyl chains are detected in the ranges of δ 1.85–1.20 and 0.87–0.85 ppm. Although the proton spectrum of M2 is complex, the ratio of the integrated intensity between the aromatic and aliphatic parts of the spectrum is consistent with the proposed structure. In the ¹³C spectrum of M2 there are 16 signals in the downfield region belonging to 16 different aromatic carbon atoms and at δ 55.3 and 14.3 ppm there are signals that are characteristic of the cyclopentane moiety in fluorene and the terminal CH₃ groups of alkyl chains, respectively. In the interval of δ 40.4–22.8 ppm signals relating to the other aliphatic carbon atoms were detected, which confirms the proposed structure (Fig. S13, ESI†).

The copolymers F1 and F2 were prepared via a Pd-mediated Stille coupling reaction between the bis(trimethyltin)-4,5-diundecylbenzo[2,1-b:3,4-b′]dithiophene monomer and the brominated monomers (M1 or M2), with yields ranging from 88% to 79% (Scheme 2).

Scheme 2 Synthesis by Stille coupling of F1 and F2 copolymers.

After polymerization, the crude polymers were washed using Soxhlet extraction with methanol, hexane and chloroform in sequence. The title polymers were obtained by reprecipitation of their concentrated solutions in chloroform from methanol. Both F1 and F2 can be readily dissolved in common solvents such as chloroform, chlorobenzene, and tetrahydrofuran, which is attributed to the incorporation of solubilizing alkyl chains joined to the thiophene flank of the polymer backbone. The structures of the obtained copolymers were confirmed by elemental analysis, ¹H NMR and UV-vis-NIR spectroscopy. The ¹H NMR spectra of F1 and F2 are shown in Fig. S14 (ESI†). The peaks at δ 7.30–9.60 ppm are assigned to the proton responses of aromatic units. There is a peak at δ 3.40–3.50 ppm, which is attributed to methylene groups. The peaks in the range of δ 1.90–0.77 ppm arise from alkyl substituents. All the ratios of integrated peak areas between the aromatic and aliphatic signals agree with the corresponding molecular structures of the polymers.

The molecular weight and polydispersity index (PDI) of the polymers were measured by gel permeation chromatography (GPC) using polystyrene as the standard and THF as the eluent. The GPC results indicated that these copolymers have a number-average molecular weight (M_n) of 14.2 and 8.5 kDa with a narrow PDI of 1.73 and 1.63 for F1 and F2, respectively. F2 has a low value of M_n, which is probably due to steric hindrance resulting from the highly rigid planar structure of the acceptor units. The thermal properties of the polymers were characterized by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) (Fig. S15, ESI†). DSC did not show an obvious glass transition for F1 and F2 under test conditions, which indicated the amorphous nature of these copolymers. The decomposition temperatures (T_d) (5% weight loss) were 334 °C and 377 °C for F1 and F2, respectively. This indicated that the thermal stability of the copolymers is sufficient for PSC applications.

Optical and electrochemical properties

The UV-visible optical absorption spectra of copolymers F1 and F2 in dilute solution in chloroform and thin film are displayed in Fig. 1 and the optical parameters are summarized in Table 1. As shown in Fig. 1, both F1 and F2 copolymers display two characteristic bands. The higher-energy band in the wavelength range of 350–600 nm arises from localized π–π transitions and the relatively broad and weak absorption in the range of 600–1300 nm can be attributed to intramolecular charge transfer between donor and acceptor units. The ICT absorption band of F2 is significantly red-shifted relative to that of F1, which may be attributed to the increased coplanarity of the backbone leading to enhanced intramolecular charge transfer in a solution of F2 after introducing fused thiazole rings into F2. Compared with F1, F2 with conjugated fused thiazole could display a further increase in spectral absorption at longer wavelengths. This phenomenon reveals that larger conjugated fused acceptor units could reduce the band gap of copolymers. In the thin film, the absorption band corresponds to a lower-energy band, which is significantly red-shifted and broadened compared with the absorption band in the solution, which is attributed to the strong intermolecular interactions and close packing of the polymer backbone in the solid state. The optical band gaps of F1 and F2 are 1.24 and 1.08 eV, respectively, as determined by the onset of absorption in the thin film. The fused heteroaromatic units in the backbone of F2 give rise to a broader absorption and narrower band gap than in F1. These results indicate that the optical band gap of copolymers depends on the electron-withdrawing ability of the polymer backbone. These results demonstrate that the strong acceptor TDQ is favorable for developing narrow-band-gap copolymers and that different combinations of donors and TDQ cores enable efficient tuning of the optical properties. Compared with other reported TDQ-based polymers, F1 and F2 displayed narrower band gaps than the values reported in the literature on TDQ-containing alternating polymers with thiophene,^5,24 fluorene^24b,25 and carbazole.^24b,26 TDQ-based polymers all display extended absorption towards the near-infrared region (beyond 1100 nm). However, both polymers display weak absorption in the region of 599–700 nm of the active solar spectrum, which may result in low photon harvesting in PSCs. The use of an organic photovoltaic material that absorbs in the NIR spectral region and even beyond 1000 nm could increase the PCE of organic BHJ solar cells.

Fig. 1 Optical absorption spectra of F1 and F2 in chloroform solution and thin film cast from chloroform.

Table 1 Optical and electrochemical properties of F1 and F2^c

Copolymer	λmax^a	λmax^b	E^optg (eV)	E^oxonset (V)	E^redonset (V)	EHOMO (eV)	ELUMO (eV)	E^elecg (eV)
a In chloroform solution.b Thin film cast from chloroform.c EHOMO/ELUMO = −(E^ox/redonset + 4.40) eV, E^optg = 1240/λonset(film).
F1	406, 844	418, 956	1.24	0.96	−0.56	−5.36	−3.84	1.52
F2	426, 964	430, 1026	1.08	1.00	−0.44	−5.40	−3.96	1.44

---

Cyclic voltammetry (CV) was employed to investigate the electrochemical properties and determine the HOMO and LUMO energy levels. The experimentally measured cyclic voltammograms are shown in Fig. 2 and the results are summarized in Table 1. The HOMO and LUMO energy levels of the copolymers were estimated according to the expressions:

HOMO = −(Eoxdonset − EFc/Fc+1/2 + 4.8) (eV)

LUMO = −(Eredonset − EFc/Fc+1/2 + 4.8) (eV)

E^ec_g = HOMO − LUMO (eV)

where the potential of ferrocene (−EFc/Fc+1/2 = 0.40 eV) vs. SCE was used as an internal standard. The values of Eoxdonset were observed to be 0.96 and 1.00 V for F1 and F2, respectively. Accordingly, the HOMO energy levels were calculated to be −5.36 and −5.40 eV for F1 and F2, which are in good agreement with the ideal HOMO energy levels for ensuring high air stability and high open-circuit voltage (Voc) in PSCs.27 Low HOMO energy levels of copolymers are desired to achieve a higher value of Voc in BHJ PSCs and to make these copolymers promising candidates for use as polymer donor materials. The HOMO level of F1 (−5.36 eV) was found to be 0.04 eV higher than that of F2 (−5.40 eV). The values of Eredonset were observed to be −0.56 and −0.44 V for F1 and F2, respectively. The LUMO levels were thus calculated to be −3.84 eV and −3.96 eV for F1 and F2, respectively, which are higher than that of PC71BM (−4.2 eV) to guarantee energetically favorable electron transfer. F2 has a lower-lying LUMO level owing to the greater electron deficiency of the benzodithiazole substituent. These results suggest that adjustment of the substitution positions and architectures has a weak influence on the HOMO and LUMO of D–A copolymers. This result shows that extension of the π-conjugation in TDQ is a viable strategy for obtaining stronger acceptors with a lower band gap of the polymers, which thus possess low LUMO and HOMO levels. The electrochemical band gaps (Eelcg) are 1.52 and 1.44 eV for F1 and F2, respectively. The difference between the optically and electrochemically measured band gaps can be explained via the exciton binding energy of the conjugated copolymers28 and the fact that electrons and holes remain electrostatically bound to one another in the excited state.29 The introduction of two donor units into F2 leads to a reduction of its LUMO level relative to that of F1 and, as a consequence, the LUMO offset (ΔELL) between F2 and PC71BM is only 0.24 eV, which is significantly less than the threshold of ∼0.3 eV for charge separation. This problem often occurs in conjugated polymers with very low optical band gaps.30 However, copolymer F1 with a D–A structure exhibits a higher LUMO energy level to give a value of ΔELL of about 0.36 eV, which is significantly higher than the threshold value for charge separation. Recently, it was reported that charge separation also occurs when the LUMO offset is below 0.3 eV,31 which indicates that F2 can also potentially be used as an electron donor for polymer solar cells.

Fig. 2 Cyclic voltammograms of films of copolymers F1 and F2 cast on a platinum electrode in 0.1 mol L⁻¹ Bu₄NClO₄/CH₃CN at a scan rate of 50 mV s⁻¹.

Theoretical calculations

We additionally performed a theoretical study on the molecular structures of F1 and F2 within the framework of density functional theory (DFT) and time-dependent density functional theory (TD-DFT). The initial geometry optimization calculations were performed employing the gradient-corrected functional PBE³² of Perdew, Burke and Ernzerhof. The def-SVP basis set³³ was used for all calculations. At this stage of the calculations, to increase the computational efficiency (without loss of accuracy) the resolution of identity method³⁴ was used for the treatment of two-electron integrals. Subsequent geometry optimization was further performed using the hybrid exchange–correlation functional B3LYP³⁵ as well as Truhlar's meta-hybrid exchange–correlation functional M06 (ref. 36) with the same basis set. Tight convergence criteria were used for the SCF energy (up to 10⁻⁷ Eh) and the one-electron density (rms of the density matrix up to 10⁻⁸) as well as for the norm of the Cartesian gradient (both average and maximum residual forces smaller than 1.5 × 10⁻⁵ a.u.) and residual displacements (both average and maximum smaller than 6 × 10⁻⁵ a.u.). Solvent effects were included for chloroform (CF) using the integral equation formalism variant of the polarizable continuum model (IEFPCM), as implemented in the Gaussian package.³⁷ TD-DFT excited-state calculations were performed to calculate the optical band gaps of F1 and F2 using the same functionals and basis set as on the corresponding ground-state structures. The UV/vis spectra were calculated using the B3LYP and M06 functionals. The first round of geometry optimization was performed using the Turbomole package.³⁸ All follow-up calculations were performed using the Gaussian package.³⁷ The first round of calculations was the geometry optimization of the structures of F1 and F2. To increase the computational efficiency, the alkyl groups were truncated to ethyl groups. Vibrational analysis of all the optimized structures did not reveal any vibrational modes with imaginary eigen frequencies, i.e., the final optimized structures are true local (if not global) minima.

In the structure of F1 the TDQ and benzodithiophene (BDT) moieties (along which the polymeric chains grow) form a planar configuration with small dihedral angles (up to 6°). The fluorene (FL) moieties form dihedral angles with TDQ in the range of 37–44° (depending on the functional used and the presence of a solvent). For the structure of F2 the thiadiazolodithiazolophenazine (TDTAP) moieties form small dihedral angles (up to 5°) with the linking thiophenes. Steric effects lead to moderate dihedral angles between TDTAP and BDT of up to 27°. We calculated the HOMO and LUMO energy levels and the optical band gaps, defined here as the energetically lowest allowed vertical electronic excitation, employing the PBE, M06, and B3LYP functionals. In Table 2, in addition to the energy levels of the frontier orbitals, we also provide the optical band gap and the main contributions to the first excitation, as well as the wavelength of the first excitation and of the excitations with the largest oscillator strengths.

Table 2 Calculated properties of F1 and F2, specifically, the HOMO and LUMO energies (eV), HOMO–LUMO gap (eV), HL, optical band gap (eV), OG, with the corresponding oscillator strengths, f, wavelengths of the first excitation and excitations with the largest oscillator strengths, main contributions to the first excited state, and dipole moment (D), μ

	HOMO (eV)	LUMO (eV)	HL (eV)	OG (eV)	λ1st/max (nm)	f	Main contributions	μ (D)
a Values when solvent effects are taken into account for chloroform.
F1
PBE	−4.75	−3.67	1.08	1.46	846	0.20	H → L (93%), H−2 → L (5%)	3.21
	−4.90^a	−3.79^a	1.11^a	1.41^a	882^a	0.04^a	H−1 → L (84%), H → L (15%)^a	4.13^a
B3LYP	−5.33	−3.23	2.10	1.82	681/419/366/327/316/295	0.25	H → L (99%)	2.79
	−5.48^a	−3.34^a	2.14^a	1.80^a	687/432/361/329/319/295^a	0.32^a	H → L (99%)^a	3.45^a
M06	−5.60	−3.17	2.44	1.89	658/398/350/322/307/292	0.24	H → L (99%)	3.00
	−5.77^a	−3.28^a	2.49^a	1.88^a	660/508/404/321/306/296^a	0.31^a	H → L (99%)^a	3.83^a
F2
PBE	−4.28	−3.75	0.53	0.92	1353	0.23	H → L (100%)	3.37
	−4.47^a	−3.92^a	0.55^a	0.89^a	1392^a	0.33^a	H → L (100%)^a	4.87^a
B3LYP	−4.80	−3.42	1.38	1.19	1043/435/400/350/337/320	0.33	H → L (100%)	2.95
	−4.99^a	−3.57^a	1.42^a	1.17^a	1063/437/407/352/327/320^a	0.45^a	H → L (100%)^a	3.89^a
M06	−5.07	−3.38	1.69	1.24	1000/419/388/335/320/308	0.34	H → L (100%)	3.23
	−5.27^a	−3.54^a	1.73^a	1.22^a	1015/420/344/336/319/309^a	0.46^a	H → L (100%)^a	4.44^a

---

In addition to the B3LYP functional, we also performed our calculations employing the M06 functional. The M06 meta-hybrid functional was chosen because it provides even performance over all transition types.^39,40 We provide results using all three functionals, which can additionally be used for comparison with the literature. The HOMO–LUMO (HL) gap for each structure calculated using the hybrid B3LYP functional is slightly smaller than that using the M06 meta-hybrid functional (by ∼0.3 eV); however, the calculated optical band gaps are only marginally smaller. All the calculated optical band gaps are low (less than 2 eV), with the lowest corresponding to the F2 structure, which also exhibits the larger oscillator strengths. In Table 2, we also provide the character of the first allowed excitations for contributions that are larger than 4% only. The first excitation, as calculated by each of the functionals for both structures, clearly exhibits a single-configuration character.

In Fig. 3, we have plotted the isosurfaces (isovalue = 0.02) of the HOMO and LUMO for both structures. For both structures, the HOMO is extended over the main body of linked groups that form the polymeric chain. In the case of F1, there is minimal delocalization over the FL moieties. In the case of F2, there is minimal delocalization over the benzobisthiazole moiety of TDTAP. The LUMO of each structure displays a higher degree of localization; for F1, the LUMO extends over TDQ and, for F2, over TDTAP. Detailed contributions of transitions to electronic excitations are given in the ESI† for all structures. To quantify the contributions of the moieties to the frontier orbitals we have calculated the total and partial density of states (PDOS). The PDOSs for F1 and F2 are shown in Fig. S16 (ESI†). We partition the structure of F1 into TDQ, BDT, FL, and alkyl moieties, and the structure of F2 into TDTAP, BDT, linking thiophenes and alkyl moieties. The two structures display similar contributions to the HOMO from the respective characteristic moieties; specifically, for F1, the contribution from the TDQ moiety is 34.7% and, for F2, the contribution from TDTAP is 35.2%. However, the contribution of BDT to the HOMO, which is 61.4%, is much higher compared with the case of F2, where it is 18.4%. For F2, the linking thiophenes contribute significantly to the HOMO, at 45.0%. The LUMO of both structures has dominant contributions from the characteristic moieties; specifically, for F1, TDQ contributes 83.3% and, for F2, the TDTAP moiety contributes 81.0%. The secondary contribution to the LUMO, for F1, is from the BDT moiety at 10.5% and, for F2, from the thiophene moieties at 15.6%. These data are in agreement with our earlier observations on the orbital delocalizations.

Fig. 3 Frontier orbitals of F1 and F2 copolymers.

In Fig. 4, we show the UV/visible absorption spectra of the structures of F1 and F2 calculated at the TD-DFT/M06 level of theory, both accounting for solvent effects of CF and in the gas phase. The spectra were produced by convoluting Gaussian functions with HWHM = 0.22 eV centered on the excitation wavenumbers. In Fig. S17 (see ESI†), we also provide the corresponding spectra calculated using the B3LYP functional, which only slightly overestimates the wavelengths (by ∼15 nm in the low-wavelength region and by ∼50 nm in the high-wavelength region) and retains all the main characteristics of those calculated using M06 and the experimental spectra. The absorption spectra of F1 and F2 exhibit two and three bands with high absorbance, respectively. For F1, the two bands are centered at 660 nm and 300 nm, with low-intensity absorption peaks between the two main bands. For F2, the three bands are located at 1015 nm and in the low-wavelength region at ∼420 nm and ∼320 nm. In the region between low and high wavelengths (500–650 nm) there is an obvious lack of absorbance even of moderate intensity.

Fig. 4 Theoretical UV/vis absorption spectra of (a) F1 and (b) F2 (calculated using the M06 functional).

Photovoltaic properties of F1 and F2

In order to balance the absorbance and charge-transporting network of the photoactive layer, the weight ratio of copolymer to PC₇₁BM was varied from 1 : 0.5 to 1 : 2.5 and a ratio of 1 : 2 provided the best photovoltaic performance for both F1 and F2-based photovoltaic devices. A further increase in the PC₇₁BM content in the active layer had a negative effect on both J_sc and FF, which resulted in a decrease in the overall PCE. Initially, we used chloroform as the solvent for solution casting of the active layer. The current–voltage characteristics under illumination and incident photon-to-current conversion efficiency (IPCE) spectra of solar cells based on optimized F1 : PC₇₁BM and F2 : PC₇₁BM with a weight ratio of 1 : 2 are presented in Fig. 5a and the corresponding photovoltaic parameters are compiled in Table 3. A polymer solar cell based on F1:PC₇₁BM displayed an overall PCE of 2.71% with J_sc = 8.36 mA cm⁻², V_oc = 0.60 V and FF = 0.54. In contrast, a device based on optimized F2:PC₇₁BM exhibited a relatively low PCE of 1.68% with a J_sc of 5.16 mA cm⁻², a V_oc of 0.64 V and a FF of 0.51. Although both the copolymers possess lower HOMO energy levels, the LUMO energy levels that are close to PC₇₁BM in energy may lead to a loss in V_oc,⁴¹ resulting in low values of V_oc. The lower value of PCE for the device based on F2:PC₇₁BM might be mainly due to the low value of J_sc, which is attributed to the insufficient exciton dissociation efficiency, as evidenced by the low value of ΔE_LL. The device based on F2:PC₇₁BM displayed a V_oc of 0.64 V, which is attributed to the relatively lower HOMO energy level of F2 compared with that of F1. The photon energy loss E_loss = E^opt_g − qV_oc equals 0.44 eV, which is below the threshold value of 0.6 eV for efficient charge transport⁴² and results in a low value of J_sc. On the other hand, the device based on F1:PC₇₁BM displayed a higher PCE with J_sc = 8.36 mA cm⁻², but a lower V_oc (0.60 V) as a consequence of the relatively higher HOMO energy level and a photon energy loss of 0.64 eV, which is higher than the threshold value. The difference in the values of J_sc is also reflected in the IPCE spectra (Fig. 5b) of the devices. Both devices display a broad IPCE response from 350 to 1050 nm and 1200 nm for F1 and F2, respectively, which covers the absorption spectra of PC₇₁BM and the copolymer. The IPCE spectra of these devices closely resemble the absorption spectra of the corresponding active layers (Fig. 6), which indicates that both PC₇₁BM and the copolymers contribute to the generation of photocurrent. As can be seen from Fig. 5b, the F2 copolymer displays relatively low IPCE values (maximum 20.4%) compared with F1 (maximum 26%). The values of J_sc estimated from the integration of the IPCE spectra of the devices based on F1:PC₇₁BM and F2:PC₇₁BM are 8.25 mA cm⁻² and 5.07 mA cm⁻², respectively, and are consistent with the values observed in the J–V characteristics under illumination. These results indicate that exciton dissociation is more efficient in F1:PC₇₁BM compared with F2:PC₇₁BM, in spite of the IPCE spectra of the latter being broader compared with those of the former.

Fig. 5 (a) Current–voltage (J–V) characteristics under illumination and (b) IPCE spectra of PSCs based on active layers processed with and without a solvent additive (SA).

Table 3 Photovoltaic parameters of BHJ organic solar cells based on F1 and F2 as the donor and PC₇₁BM as the acceptor processed with CF and DIO/CF

Active layer	Jsc (mA cm^−2)	Voc (V)	FF	PCE (%)	PCE^c (%) (average)
a Cast from CF.b Cast from DIO/CF.c Average of 10 devices.
F1:PC71BM^a	8.36	0.60	0.54	2.71	2.65
F2:PC71BM^a	5.16	0.64	0.51	1.68	1.60
F1:PC71BM^b	14.71	0.58	0.68	5.80	5.74
F2:PC71BM^b	8.92	0.60	0.62	3.32	3.24

---

Fig. 6 Absorption spectra of blended thin films of F1:PC₇₁BM and F2:PC₇₁BM with and without a solvent additive (SA).

Here, we tried to increase the PCE of the devices by employing a solvent additive method to prepare the active layer.⁴³ 1,8-Diiodooctane (DIO) was added as a processing additive to study its effect on the overall PCE of the devices. For the solvent additive treatment, we varied the concentration of DIO from 0.5 to 4.5% and found that a concentration of 4% DIO in the host CF solution provides the best photovoltaic performance. The J–V characteristics under illumination and corresponding IPCE spectra of the devices are shown in Fig. 5a and b, respectively. The overall PCE of the devices based on F1:PC₇₁BM and F2:PC₇₁BM was increased to 5.79% (J_sc = 14.71 mA cm⁻², V_oc = 0.58 V and FF = 0.68) and 3.32% (J_sc = 8.92 mA cm⁻², V_oc = 0.60 V and FF = 0.62), respectively. The increase in PCE is associated with increases in J_sc and FF with a slight decline in V_oc. The increases in J_sc and FF are mainly because DIO possesses a high boiling point and is able to solvate PC₇₁BM, leading to a significant effect on the morphology, which facilitates charge carrier transportation.⁴⁴ However, the slight decline in V_oc can be attributed to a reduction in charge separated and charge transfer state energies upon the incorporation of DIO.⁴⁵ The addition of DIO significantly improved the performance of the BHJ polymer solar cells by reducing the size, increasing the D–A interfacial area within the active layer⁴⁶ and helping to increase the values of J_sc and FF. The values of IPCE also increased significantly for the devices based on an active layer cast using DIO/CF (Fig. 5b). The values of J_sc estimated from the integration of the IPCE spectra of the devices based on F1 and F2 are 14.62 mA cm⁻² and 8.76 mA cm⁻², respectively.

The PCE of a PSC depends upon the light-harvesting ability of the device, which is directly related to its molar absorption coefficient and response. In order to obtain information about the effect of the solvent additive on the absorption properties, the absorption spectra of the active layer processed with and without a solvent additive were recorded and are shown in Fig. 6. As shown in Fig. 6, in comparison with the spectra of films of the active layer cast from CF, in the spectra of films cast from DIO/CF the absorption peaks that correspond to the ICT band of the copolymers are red-shifted and also the absorption intensity was increased. These effects are related to enhanced π–π stacking. The increase in the absorption of the active layer produces more excitons, thereby resulting in a higher value of J_sc.

In order to obtain information about the increased PCE of the PSCs, the morphology of the active layer was investigated using transmission electron microscopy (TEM). As shown in Fig. 7, the blended films of copolymer:PC₇₁BM without SA display bigger domains. This morphology may be a reason for the insufficient exciton diffusion and dissociation and may be associated with the low values of J_sc. The morphology of the active layer that was spin-cast with SA was different. The large crystallites and phase-separated domains were reduced, resulting in favorable exciton diffusion and dissociation and charge transport. Among all the films, F1:PC₇₁BM cast with SA displayed a much finer and denser texture and nanophase separation between the polymer and PC₇₁BM, which could result in a high PCE with a high value of J_sc.⁴⁷

Fig. 7 TEM images of F2:PC₇₁BM (a and b) and F1:PC₇₁BM (c and d) films, without (a and c) and with (b and d) a solvent additive (SA). Scale bar = 200 nm.

Besides the absorption and energy levels, the charge carrier transport ability within the active layer is also a critical factor for the resulting photovoltaic performance of a PSC, especially the FF. We measured the hole mobility in the active layers processed with and without DIO by the space charge limited current (SCLC) method⁴⁸ using hole-only devices. The SCLC could be estimated using the Mott–Gurney expression: J = (9/8) ε_oε_rμ(V²/d³), where J is the current density, ε_r is the dielectric constant of the active layer, ε_o is the permittivity of free space (8.85 × 10⁻¹² F m⁻¹), d is the thickness of the active layer (90 nm), V (=V_appl − V_bi) is the effective voltage, V_appl is the applied voltage, and V_bi is the built-in voltage that arises from the difference in work function between the anode and cathode. The J–V characteristics of hole-only devices (ITO/PEDOT:PSS/active layer/Au) for optimized active layers of F1:PC₇₁BM and F2:PC₇₁BM processed with and without SA are shown in Fig. 8. The hole mobilities estimated from Fig. 8 by fitting with the Mott–Gurney equation are 3.67 × 10⁻⁵ cm² V⁻¹ s⁻¹ and 1.86 × 10⁻⁵ cm² V⁻¹ s⁻¹ for F1:PC₇₁BM and F2:PC₇₁BM, respectively, spin-cast from CF only. When the active layer was deposited using SA, the hole mobility increased to 8.76 × 10⁻⁵ cm² V⁻¹ s⁻¹ and 6.75 × 10⁻⁵ cm² V⁻¹ s⁻¹ for F1:PC₇₁BM and F2:PC₇₁BM, respectively. The electron mobilities were also estimated using electron-only devices (ITO/Al/active layer/Al) for optimized active layers of F1:PC₇₁BM and F2:PC₇₁BM processed with and without SA using SCLC J–V characteristics in the dark (as shown in Fig. 9 for devices based on F1:PC₇₁BM only) and are 2.34 × 10⁻⁴ cm² V⁻¹ s⁻¹, 2.42 × 10⁻⁴ cm² V⁻¹ s⁻¹, 2.44 × 10⁻⁴ cm² V⁻¹ s⁻¹ and 2.48 × 10⁻⁴ cm² V⁻¹ s⁻¹ for F1:PC₇₁BM (as cast), F1:PC₇₁BM (SA), F2:PC₇₁BM (as cast) and F2:PC₇₁BM (SA), respectively. The higher hole mobility and slight change in electron mobility result in more balanced charge transport, which is beneficial for an improvement in the FF.⁴⁷

Fig. 8 J–V characteristics of hole-only devices based on blended thin films of F1:PC₇₁BM and F2:PC₇₁BM processed with and without SA. Lines represent SCLC fitting.

Fig. 9 J–V characteristics of electron-only devices based on blended thin films of F1:PC₇₁BM processed with and without SA. Lines represent SCLC fitting.

Conclusions

In summary, two D–A conjugated copolymers F1 and F2 with ultra-low band gaps of 1.24 eV and 1.08 eV were designed and applied as donors along with PC₇₁BM as an electron acceptor for solution-processed polymer solar cells. After the optimization of the active layer, F1:PC₇₁BM and F2:PC₇₁BM processed with a 4% DIO/CF (SA) solution displayed an overall PCE of 5.80% and 3.32%, respectively. The device based on F2 displayed a relatively low PCE compared with that of F1, in spite of its low optical band gap and broader absorption profile relative to F1, which is mainly attributed to an insufficient driving force for exciton dissociation owing to the low value of the LUMO offset between F2 and PC₇₁BM and also because of the low photon energy loss, which is below the threshold value of 0.6 eV for efficient charge transport. Because the HOMO energy level of both copolymers is quite low, there is an energy barrier with PEDOT:PSS for hole extraction, which limits the value of J_sc. This can be overcome by employing a hole-transporting layer with a high work function that matches the HOMO energy level of the copolymer. This work is in progress and will be reported on when completed. The above investigation also demonstrates that these copolymers have the potential in ternary BHJ and tandem solar cells to achieve a high PCE. Research work in this direction is in progress.

Acknowledgements

M. L. K., S. A. K., N. A. R., A. Y. N. and G. D. S. are thankful to the Russian Science Foundation (grant number 14-13-01444) for financial assistance.

References

1. (a) Q. Gan, F. J. Bartoli and Z. H. Kafafi, Plasmonic-Enhanced Organic Photovoltaics: Breaking the 10% Efficiency Barrier, Adv. Mater., 2013, 25, 2385–2396; (b) A. J. Heeger, 25^th Anniversary Article: Bulk Heterojunction Solar Cells: Understanding the Mechanism of Operation, Adv. Mater., 2014, 26, 10–28; (c) Z. He, H. Wu and Y. Cao, Recent Advances in Polymer Solar Cells: Realization of High Device Performance by Incorporating Water/Alcohol Soluble Conjugated Polymers as Electrode Buffer Layer, Adv. Mater., 2014, 26, 1006–1024; (d) L. Dou, J. You, Z. Hong, Z. Xu, G. Li, R. A. Street and Y. Yang, 25^th Anniversary Article: A Decade of Organic/Polymeric Photovoltaic Research, Adv. Mater., 2013, 25, 6642–6671; (e) F. Machui, M. Hosel, N. Li, G. D. Spyropoulos, T. Ameri, R. R. Sondergaard, M. Jorgensen, A. Scheel, D. Gaiser, K. Kreul, D. Lenssen, M. Legros, N. Lemaitre, M. Vilkman, M. Valimaki, S. Nordman, C. J. Brabec and F. C. Krebs, Cost Analysis of Roll to Roll Fabricated ITO Free Single and Tandem Organic Solar Modules Based Data From Manufacture, Energy Environ. Sci., 2014, 7, 2792–2802; (f) P.-L. T. Boudreault, A. Najari and M. Leclerc, Proceessable Low Bandgap Polymers for Photovoltaic Applications, Chem. Mater., 2011, 23, 456–469; (g) H. J. Son, B. Carsten, I. H. Jung and L. Yu, Overcoming Efficiency Challenges in Organic Solar Cells: Rational Development of Conjugated Polymers, Energy Environ. Sci., 2012, 5, 8158–8170; (h) L. Lu, T. Zheng, Q. Wu, A. M. Schneider, D. Zhao, L. Yu, Recent Advances in Bulk Heterojunction Polymer Solar Cells, Chem. Rev., 2015, 115, 12666–12731; (i) G. Li, R. Zhu and Y. Yang, Polymer Solar Cells, Nat. Photonics, 2012, 6, 153–161.
2. (a) Y. Liu, J. Zhao, Z. Li, C. Mu, W. Ma, H. Hu, K. Jiang, H. Lin, H. Ade and H. Yan, Aggregation and Morphology Control Enables Multiple Cases of High Efficiency Polymer Solar Cells, Nat. Commun., 2014, 5, 5293–5530; (b) Z. He, B. Xiao, F. Liu, H. Wu, Y. Yang, S. Xiao, C. Wang, T. P. Russell and Y. Cao, single Junction Polymer Solar Cells With High Efficiency and Photovoltage, Nat. Photonics, 2015, 9, 174–179; (c) M. A. Green, K. Emery, Y. Hishikawa, W. Warta and E. D. Dunlop, Solar Cells Efficiency Tables (Version 45), Prog. Photovoltaics, 2015, 23, 1–9; (d) C. Liu, C. Yi, K. Wang, Y. Yang, R. S. Bhatta, M. Tsige, S. Xiao and X. Gong, Single Junction Polymer Solar Cells Over 10% Efficiency by a Novel Two Dimensional Donor–Acceptor Conjugated Copolymers, ACS Appl. Mater. Interfaces, 2015, 7, 4928–4935; (e) S.-H. Liao, H.-J. Jhuo, Y.-S. Cheng and S.-A. Chen, Fullerene Derivative-Doped Zinc Oxide Nanofilm as the Cathode of Inverted Polymer Solar Cells With Low bandgap Polymer PTB7-Th for High Performance, Adv. Mater., 2013, 25, 4766–4771; (f) S.-H. Liao, H.-J. Jhuo, P.-N. Yeh, Y.-S. Cheng, Y.-L. Li, Y.-H. Lee, S. Sharma and S.-A. Chen, Single Junction Inverted Polymer Solar Cells Reaching Power Conversion Efficiency 10.31% by Employing Dual Doped Zinc Oxide Nano-film as Cathode Interlayer, Sci. Rep., 2014, 4, 6813; (g) J. D. Chen, C. Cui, Y. Q. Li, Z. Zhou, Q. D. Cu, C. Li, Y. Li and Z. Tang, Single Junction Polymer Solar Cells Exceeding 10% Power Conversion Efficiency, Adv. Mater., 2015, 27, 1035–1041.
3. M. Scharber and N. Sariciftci, Efficiency of Bulk Heterojunction Organic Solar Cells, Prog. Polym. Sci., 2013, 38, 1929–1940.
4. H. Choi, S. J. Ko, J. Kim, P. Q. Morin, B. Walker, B. H. Lee, M. Leclerc, Y. M. Kim and A. J. Heeger, Small Bandgap Polymer Solar Cells With Unexpected Short Circuit Current Density and High Fill Factor, Adv. Mater., 2015, 27, 3318–3324.
5. E. Zhou, K. Hashimoto and K. Tajima, Low Bandgap Polymers for Photovoltaic Device With Photocurrent Response Wavelengths Over 1000 nm, Polymer, 2013, 54, 6501–6509.
6. E. E. Havinga, W. ten Hoeve and H. Wynberg, A New Class of Small bandgap Polymer Conductors, Polym. Bull., 1992, 29, 119–126; P. M. Beaujuge, C. M. Amb and J. R. Reynolds, Spectral Engineering in π-Conjugated Polymers With Intramolecular Donor–Acceptor Interactions, Acc. Chem. Res., 2010, 43, 1396–1407.
7. (a) E. Perzon, F. Zhang, M. Andersson, W. Mammo, O. A. Inganas and M. R. Andersson, A Conjugated Polymer for Near Infrared Optoelectronic Applications, Adv. Mater., 2007, 19, 3308–3311; (b) K. G. Jespersen, W. J. D. Beenken, Y. Zaushitsyn, A. Yartsev, M. Andersson, T. Pullerits and V. Sundstrom, The Electronic State of Polyfluorene Copolymers With Alternating Donor–Acceptor Units, J. Chem. Phys., 2004, 121, 12613–12617.
8. (a) K. H. Hendriks, G. H. Heintges, V. S. Gevaerts, M. M. Wienk and R. A. J. Janssen, High Molecular Weight Regular Alternating Diketopyrrolopyrrole Based Terpolymers for Efficient Organic Solar Cells, Angew. Chem., Int. Ed., 2013, 52, 8341–8344; (b) R. S. Ashaf, I. Meager, M. Nikolka, M. Kirkers, M. Planells, B. C. Schroeder, S. Holliday, M. Hurhangee, C. B. Nielsen, H. Sirringhaus and I. McCulloch, Chalcogenophene Comonomers Comparison in Small Bandgap Diketopyrrolopyrrole Based Conjugated Polymers for High Performing Field Effect Transistors and Organic Solar Cells, J. Am. Chem. Soc., 2015, 137, 1314–1321; (c) C. B. Nielsen, R. S. Ashaf, N. D. Treat, B. C. Schroeder, J. E. Donaghey, A. J. White and I. McCulloch, 2,1,3-Benzothiadiazole-5,6-Dicarboxylic Imide – A Versatile Building Block for Additive and Annealing Free Processing or Organic Solar Cells With Efficiencies Exceeding 8%, Adv. Mater., 2015, 27, 948–953; (d) Z. Zhang, F. Lin, H. C. Chen, H. C. Wu, C. L. Chung, C. Lu, S. H. Liu, S. H. Tung, W. C. Chen, K. T. Wong and P. T. Chou, A Silole Copolymer Containing a Ladder Type Heptacylic arene and Naphthobisoxadiazole Moieties for Highly Efficient Polymer Solar Cells, Energy Environ. Sci., 2015, 8, 552–557.
9. (a) D. H. Wang, A. K. K. Kyaw, J. R. Pouliot, M. Leclerc, A. J. Heeger, Enhanced Power Conversion Efficiency of Low Bandgap Polymer Solar Cells by Insertion of Optimized Binary Processing Additives, Adv. Energy Mater., 2014, 4, 201300835; (b) J. Adams, G. D. Spyropoulos, M. Salvador, N. Li, S. Strohm, L. Lucera, S. Langner, F. Machui, H. Zhang, T. Ameri, M. M. Voiget, F. c. Frebs and C. J. Brabec, Air Processed Organic Tandem Solar Cells on Glass: Toward Competitive Operating Lifetimes, Energy Environ. Sci., 2015, 8, 169–176; (c) T. Mo, K. Jiang, S. Chen, H. Hu, H. Lin, Z. Lin, J. Zhao, Y. Liu, Y. M. Chang, C. C. Hsiao, H. Yan, Efficient Low Bandgap Polymer Solar Cells With High Open Circuit Voltage and Good Stability, Adv. Energy Mater., 2015, 5, 201501282.
10. (a) U. Wurfel, D. Nehar, A. Spies and A. Albrecht, Impact of Charge Transport on Current–Voltage Characteristics and Power Conversion Efficiency of Organic Solar Cells, Nat. Commun., 2015, 6, 6951; (b) D. Bartesaghi, C. Perez Ideal, J. Kniepert, S. Roland, M. Turbiez, D. Neher and L. J. Koster, Nat. Commun., 2015, 6, 7083.
11. J. W. Jung, F. Liu, T. P. Russell and W. H. Jo, A High Mobility Conjugated Polymer Based on Dithienothiophene and Diketopyrrolopyrrole for Organic Solar Cells, Energy Environ. Sci., 2012, 5, 6857–6861.
12. Science Daily, February 13, 2012. http://www.sciencedaily.com/releases/2012/02/120213133709.htm.
13. (a) X. Guo, S. R. Puniredd, B. He, T. Marszalak, M. Baumgerten, W. Pisula and K. Mullen, Combination of Two Diketopyrrolopyrrole Isomers in One Polymer for Ambipolar Transport, Chem. Mater., 2014, 26, 3595–3598; (b) W. Li, W. S. C. Roelorfs, M. Turbiez, M. M. Wienk and R. A. J. Janssen, Polymer Solar Cells With Diketopyrrolopyrrole Conjugated Polymers as the Electron Donor and Electron Acceptor, Adv. Mater., 2014, 3304–3306.
14. Y. Wnag, T. Kadoya, L. Wang, T. Hayakawa, M. Tokita, T. Mori and T. Michinobu, Benzobisthiadiazole Based Conjugated Donor–Acceptor Polymers for Organic Thin Film Transistors: Effects of π-Conjugated Bridges on Ambipolar Transport, J. Mater. Chem. C, 2015, 3, 1196–1207.
15. T. L. D. Tam, T. Selim, H. Li, F. Zhou, S. G. Mhaisalkar, H. Su, Y. M. Lam and A. C. Grimsdale, From Benzobisthiadiazole, Thaidiazoloquinoxaline to Pyrazinoquinoxaline Based Polymers: Effects of Aromatic Substituents on the Performance of Organic Photovoltaics, J. Mater. Chem., 2012, 22, 18528–18534.
16. (a) C. An, M. Li, T. Marszalek, X. Guo, W. Pisula and M. Baumgarten, Investigation of the Structure–Property Relationship of Thiadiazoloquinoxaline Based Copolymer Semiconductors Via Molecular Engineering, J. Mater. Chem. C, 2015, 3, 3876–3881; (b) T. T. Steckler, P. Henriksson, S. Mollinger, A. Lundin, A. Salleo and M. R. Andersson, Very Low Bandgap Thiadiazoloquinoxaline Donor–Acceptor Polymers as Multi-tool Conjugated Polymers, J. Am. Chem. Soc., 2014, 136, 1190–1193; (c) C. An, S. R. Puniredd, X. Guo, T. Stelzig, Y. Zhao, W. Pisula and M. Baumgarten, Benzodithiophene–Thiadiazoloquinoxaline as an Acceptor for Ambipolar Copolymers With Deep LUMO Level and Distinct Linkage Pattern, Macromolecules, 2014, 47, 979–986.
17. A. Balan, D. Baran and L. Toppare, Benzotriazole Containing Conjugated Polymers for Multi-purpose Organic Electronic Applications, Polym. Chem., 2011, 2, 1029–1043.
18. M. Karakawa and Y. Aso, Near Infrared Photovoltaic Performance of Conjugated Polymers Containing Thienoisoindigo Acceptor Units, Macromol. Chem. Phys., 2013, 21, 2388–2397.
19. K. H. Hendrinks, W. Li, M. M. Wienk and R. A. J. Janssen, Small Bandgap Semiconducting Polymers With High Near Infrared Photoresponse, J. Am. Chem. Soc., 2014, 136, 12130–12136.
20. E. Zhou, M. Nakano, S. Izawa, J. Cong, I. Osaka, K. Takimiya and K. Tajima, All Polymer Solar Cell With High Near Infrared Response Based on a Naphthodithiophene Diimide (NDTI) Copolymer, ACS Macro Lett., 2014, 3, 872–875.
21. (a) J. Hai, W. Yu, E. Zhu, L. Bian, J. Zhang and W. Tang, Thin Solid Films, 2014, 562, 75–83; (b) J. Hai, G. Shi, J. Yu, E. Zhu, Li. Bian, W. Ma and W. Tang, Napthodifuran Alternating Quinoxaline Copolymers With a Bandgap of ∼1.2 eV and Their Photovoltaic Characterization, New J. Chem., 2014, 38, 4816–4822; (c) B. M. Squeo, N. Gasparini, T. Ameri, A. P. Cando, S. Allard, V. G. Gregoriou, C. J. Brabec, U. Scherfc and C. L. Chochos, Ultra Low Bandgap a α,β-un-substituted BODIPY Based Copolymer Synthesized by Palladium Catalyzed Cross-coupling Polymerization for Near Infrared Organic Photovoltaics, J. Mater. Chem. A, 2015, 3, 16279–16286.
22. (a) T. Dallos, D. Beckmann, G. Brunklaus and M. Baumgarten, Thiadiazoloquinoxaline–Acetylene Containing Polymers as Semiconductors in Ambipolar Field Effect Transistors, J. Am. Chem. Soc., 2011, 133, 13898–13901; (b) C. An, R. Puniredd, T. Stelzzing, Y. Zhao, W. Pisula and M. Baumgarten, Impact of the Alkyl Side Chains on the Optoelectronic Properties of a Series of Photovoltaic Low Bandgap Copolymers, Macromolecules, 2014, 47, 979–986.
23. M. L. Keshtov, S. N. Osipov, M. A. Topchiy, M. A. Zotova, I. O. Konstantinov, M. M. Krayushkin, S. A. Kuklin and A. R. Khokhlov, Synthesis of New Symmetrical carbazole and Fluorene Containing α-Diketones, Dokl. Chem., 2015, 463, 215–220.
24. (a) Y. Lee, T. P. Russell and W. H. Jo, Synthesis and photovoltaic properties of low bandgap copolymers consisting of 3-hexylthiophene and [1,2,5]thiadiazolo[3,4-g]quinoxaline derivatives, Org. Electron., 2010, 11, 846–853; (b) X. Wang, E. Perzon, J. l. Delgado, P. de La Cruz, F. Zhang, F. Langa, M. Andersson and O. Inganas, Infrared photocurrent spectral response from plastic solar cell with low bandgap polyfluorene and fullerene derivatives, Appl. Phys. Lett., 2004, 85, 5081–5083.
25. H. Yi, R. G. Johnson, A. Iraqi, D. Mohamad, R. Royce and D. G. Lidzey, Narrow energy gap polymers with absorption up to 1200 nm and their photovoltaic properties, Macromol. Rapid Commun., 2008, 29, 1804–1809.
26. H. Zhou, L. Yang and W. You, Rational Design of High Performance Conjugated Polymers for Organic Solar Cells, Macromolecules, 2012, 45, 607–632.
27. N. S. Sariciftci, Primary Photoexcitations in Conjugated Polymers: Molecular Excitons vs. Semiconductor Band Model, World Scientific, Singapore, 1997.
28. (a) C. W. Schlenker and M. E. Thompson, The Molecular Nature of Photovoltage Losses in Organic Solar Cells, Chem. Commun., 2011, 47, 3702–3716; (b) W. J. Postscavage Jr, A. Sharma and B. Kippelen, Critical Interfaces in Organic Solar Cells and Their Influence on Open Circuit Voltage, Acc. Chem. Res., 2009, 42, 1758–1767.
29. (a) A. P. Zoombelt, S. G. J. Mathijssen, M. G. R. Turbiez, M. M. Wienk and R. A. J. Janssen, Small Bandgap Polymers Based on Diketopyrrolopyrrole, J. Mater. Chem., 2010, 20, 2240–2246; (b) F. L. Zhang, J. Bijleveld, E. Perzon, K. Tvingstedt, S. Barrau, O. Inganäs and M. R. Andersson, High Photovoltage Achieved in Low bandgap Polymer Solar Cells by Adjusting Energy Levels of a Polymer With the LUMOs of Fullerene Derivatives, J. Mater. Chem., 2008, 18, 5468–5474; (c) W. Li, K. H. Hendriks, A. Furlan, A. Zhang, M. M. Wienk and R. A. J. Janssen, A Regioregular Terpolymer Comprising Two Electron Deficient and One Electron Rich Unit for Ultra Small Bandgap Solar Cells, Chem. Commun., 2015, 51, 4290–4293.
30. (a) W. Li, K. H. Hendriks, A. Furlan, M. M. Wienk and R. A. J. Janssen, High Quantum Efficiencies in Polymer Solar Cells at Energy Losses Below 0.6 eV, J. Am. Chem. Soc., 2015, 137, 2231–2234; (b) K. Gao, L. Li, T. Lai, L. Xiao, Y. Huang, F. Huang, J. Peng, Y. Cao, F. Liu, T. P. Russell, R. A. J. Janssen and X. Peng, Deep Absorbing Porphyrin Small Molecule for High Performance Organic Solar Cells With Very Low Energy Losses, J. Am. Chem. Soc., 2015, 137, 7282–7285.
31. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett., 1996, 77, 3865–3868.
32. A. Schafer, H. Horn and R. Ahlrichs, Fully optimized contracted Gaussian basic sets for atoms Li to Kr, J. Chem. Phys., 1992, 97, 2571–2577.
33. K. Eichkorn, O. Treutler, H. Öhm, M. Häser and R. Ahlrichs, Auxiliary basis sets to approximate coulomb potentials, Chem. Phys. Lett., 1995, 240, 283–290.
34. (a) A. D. Becke, Density functional thermo-chemistry III, the role of exact exchange, J. Chem. Phys., 1993, 98, 5648–5652; (b) C. Lee, W. Yang and R. G. Parr, Development of the Colle–Salvetti correlation energy formula into a functional of the electron density, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789.
35. Y. Zhao and D. G. Truhlar, The MO6 suite of density functionals for main group thermo-chemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four MO6-class functionals and 12 other functionals, Theor. Chem. Acc., 2008, 120, 215–241.
36. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, 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, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 03, revision C.01, Gaussian, Inc., Wallingford CT, 2004.
37. TURBOMOLE (version 5.6), Universitat Karlsruhe, 2000.
38. D. Jacquemin, E. A. Perpète, I. Ciofini, C. Adamo, R. Valero, Y. Zhao and D. G. Truhlar, On the performance of M06 family of density functionals for electronic excitation energies, J. Chem. Theory Comput., 2010, 6, 2071–2085.
39. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Dye sensitized solar cells with 13% efficiency achieved through the molecular engineering of porphyrins sensitizers, Nat. Chem., 2014, 6, 242–247.
40. A. B. Yusoff, D. Kim, H. P. Kim, F. K. Shncider, W. J. deSilva and J. Jang, A High Efficiency Solution Processed Polymer Inverted Triple-Junction Solar Cell Exhibiting a Power Conversion Efficiency of 11.83%, Energy Environ. Sci., 2015, 8, 303–316.
41. D. Veldman, S. C. J. Meskers and R. A. J. Janssen, The Energy of Charge Transfer States in Electron Donor–Acceptor Blends: Insight Into the Energy Losses in Organic Solar Cells, Adv. Funct. Mater., 2009, 19, 1939–1948.
42. (a) W. Kim, J. K. Kim, E. Kim, T. K. Ahn, D. H. Wang and J. H. Park, Conflicted Effects of a Solvent Additive on PTB7:PC71BM Bulk Heterojunction Solar Cells, J. Phys. Chem. C, 2015, 119, 5954–5961; (b) W. Zhou, H. Chen, J. Lv, Y. Chen, W. Zhang, G. Yu and F. Li, Improving the Efficiency of Polymer Solar Cells Based on Furan-Flanked Diketopyrrolopyrrole Copolymer Via Solvent Additive and Methanol Treatment, Nanoscale, 2015, 7, 15945–15952.
43. (a) J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. Heeger and G. C. Bazan, Efficiency Enhancement in Low Bandgap Polymer Solar Cells by Processing Polymer Solar Cells by Processing With Alkane Dithiols, Nat. Mater., 2007, 6, 497–500; (b) J. K. Lee, W. L. Ma, C. J. Brabec, J. Yuen, J. S. Moon, J. Y. Kim, K. Lee, G. C. Bazan and A. Heeger, Processing Additive for Improved Efficiency From Bulk Heterojunction Solar Cells, J. Am. Chem. Soc., 2008, 130, 3619–3623.
44. (a) D. D. Nuzzo, A. Aguirre, M. Shahid, V. S. Gevaerts, S. C. J. Meskers and R. A. J. Janssen, Improved Film Morphology Reduces Charge Carrier Recombination Into the Triplet Excited State in a Small Bandgap Polymer–Fullerene Photovoltaic Cell, Adv. Mater., 2010, 22, 4321–4324; (b) Z. Qiao, M. Wang, M. Zhao, Z. G. Zhang, Y. Li, X. Li, H. Wang, Effect of Fluorine Substitution on the Photovoltaic Performance of Poly(thiophene–quinoxaline) Copolymers, Polym. Chem., 2015, 6, 8203–8213.
45. S. H. Park, A. Roy, S. Beaupré, S. Cho, N. Coates, J. S. Moon, D. Moses, M. Leclerc, K. Lee and A. J. Heeger, Bulk Heterojunction Solar Cells With Internal Quantum Efficiency Approaching 100%, Nat. Photonics, 2009, 3, 297–302.
46. (a) Y. Deng, J. Liu, J. Wang, L. Liu, W. Li, H. Tian, X. Zhang, Z. Xie, Y. Geng and F. Wang, Dithienocarbazole and Isoindigo Based Amorphous Low Bandgap Conjugated Polymers for Efficient Polymer Solar Cells, Adv. Mater., 2014, 26, 471–476; (b) J. W. Jung, F. Liu, T. P. Russell and W. H. Jo, Semi-crystalline random conjugated copolymers with panchromatic absorption for highly efficient polymer solar cells, Energy Environ. Sci., 2013, 6, 3301–3307; (c) G. E. Park, J. Shin, D. H. Lee, M. J. Cho and D. H. Choi, Effect of branched alkyl side chains on the performance of thin film transistors and photovoltaic cells fabricated with isoindigo based conjugated polymers, J. Polym. Sci., Part A: Polym. Chem., 2015, 53, 1226–1234.
47. P. W. M. Blom, V. D. Mihailetchi, L. J. A. Koster and D. E. Markov, Device Physics of Polymer: Fullerene Bulk Heterojunction Solar Cells, Adv. Mater., 2007, 19, 1551–1566.
48. (a) W. Li, S. Albrecht, L. Yang, S. Roland, J. R. Tumbleston, T. McAfee, L. Yan, M. A. Kelly, H. Ade, D. Neher and W. You, Mobility Controlled Performance of Thick Solar Cells on Fluorinated Copolymers, J. Am. Chem. Soc., 2014, 136, 15566–15576; (b) K. Li, Z. Li, X. Xu, L. Wang and Q. Peng, Development of Large Bandgap Conjugated Copolymers for Efficient Regular Single and Tandem Organic Solar Cells, J. Am. Chem. Soc., 2013, 135, 13549.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra24364e

---