Near-Infrared All-Fused-Ring Nonfullerene Acceptors Achieving an Optimal Efficiency-Cost-Stability Balance in Organic Solar Cells

Open AccessCCS ChemistryRESEARCH ARTICLE28 Apr 2022Near-Infrared All-Fused-Ring Nonfullerene Acceptors Achieving an Optimal Efficiency-Cost-Stability Balance in Organic Solar Cells Wenrui Liu†, Shengjie Xu†, Hanjian Lai†, Wuyue Liu, Feng He and Xiaozhang Zhu Wenrui Liu† Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100049 †W. Liu, S. Xu, and H. Lai contributed equally to this work.Google Scholar More articles by this author , Shengjie Xu† Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 †W. Liu, S. Xu, and H. Lai contributed equally to this work.Google Scholar More articles by this author , Hanjian Lai† Department of Chemistry, Shenzhen Grubbs Institute, Southern University of Science and Technology, Shenzhen 518055 †W. Liu, S. Xu, and H. Lai contributed equally to this work.Google Scholar More articles by this author , Wuyue Liu Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author , Feng He Department of Chemistry, Shenzhen Grubbs Institute, Southern University of Science and Technology, Shenzhen 518055 Google Scholar More articles by this author and Xiaozhang Zhu *Corresponding author: E-mail Address: [email protected] Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190 School of Chemical Sciences, University of Chinese Academy of Sciences, Beijing 100049 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202201963 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Synergistically achieving stability, cost, and efficiency is crucial for the commercialization of organic solar cells (OSCs). Despite the rapid development of 2-(3-oxo-2,3-dihydro-1H-inden-1-ylidene)malononitrile-type nonfullerene acceptors (NFAs), they are inherently unstable due to the vulnerable exocyclic double bond and possess high synthesis complexity (SC). Based on the “all-fused-ring electron acceptor (AFAR)” concept, we report two new near-infrared NFAs, F11 and F13, featuring all fused dodecacyclic rings. By developing a whole set of synthetic procedures, F11 and F13 can be conveniently prepared at a 10 g scale within a notably short period, displaying both the low SC and the lowest costs among reported NFAs, even comparable to the classical photovoltaic material, P3HT. In comparison with the one-dimensional stacking of ITYM (ITYM = 2,2′-(7,7,15,15-tetrahexyl-7,15-dihydro-s-indaceno[1,2-b:5,6-b′]diindeno[1,2-d]thiophene-2,10(2H)-diylidene)dimalononitrile), the first AFRA, and mixed J- and H-aggregations in Y6, F-acceptors show a compact honeycomb-type three-dimensional stacking with exclusive J-aggregations, favoring multichannel charge transport. By matching a medium-bandgap polymer donor, F13 delivers greater than 13% power conversion efficiencies, which is the highest performance among non-INCN acceptors, and shows device stability superior to the typical ITIC- and Y6-based OSCs as evidenced by the negligible burn-in losses. This work presents a first and successful example of NFAs achieving an optimal efficiency-cost-stability balance in OSCs. Download figure Download PowerPoint Introduction Organic solar cells (OSCs) are considered a promising device for sustainable energy because of their great potential for low cost, light weight, and large-area processability.1–5 The last few years have witnessed a great breakthrough in the power conversion efficiency (PCE) of OSCs benefiting from the innovation of materials, especially the development of nonfullerene acceptors (NFAs).6–21 Stability and cost, which are another two key issues for the commercialization of OSCs, are being paid increasing attention.14,22–25 In comparison with a fullerene-based counterpart, the NFA-based device exhibiting the optimal efficiency-cost-stability balance is a competitive choice for practical applications,14 which is critical but challenging and yet to be achieved. Acceptor–donor–acceptor (A–D–A)-type NFAs consisting of multiple conjugated heteroaromatic rings as donors (Ds) and 2-(3-oxo-2,3-dihydro-1H-inden-1-ylidene)malononitrile (INCN) as the acceptor (A) are inherently unstable because of the vulnerable exocyclic double bond, which is the main factor limiting the device lifetime of high-performance nonfullerene OSCs,26 as proved by less degradation within polymer donors.27,28 Jiang et al.27 reported the photodegradation of the high-performance NFAs (IT-4F, ITIC, and IEICO-4F) assisted by the photocatalytic properties of ZnO, leading to disruption of the C=C linkage between the D and A. Park and Son29 confirmed that ITIC reacted with hydroxyl radicals, while the ITIC radical product acted as an electrophile and thus attacked the enone group of another ITIC molecule, resulting in the cleavage of the double bond. In addition to extrinsic influence, NFAs can react with an amine-containing interfacial layer.30,31 Quite recently, it has been reported that INCN-type NFAs produced fused-ring isomers involving intramolecular six-electron electrocyclizations in the photodegradation.32 To address the stability issue, molecular engineering, considered the essential solution, was presented. Liu et al.33 identified that structural confinement by installing an outward-chain in the A–D–A NFAs suppressed the photoisomerization of vinyl groups; however, the vulnerable exocyclic double bond still exists. Moreover, replacing INCN with rhodanine can boost stability to a great extent; for example,34,35 Gasparini et al.34 presented that OSCs based on rhodanine-benzothiadiazole-coupled indacenodithiophene are free of burn-in, the commonly observed rapid performance loss. Later, Liu et al.36 presented a molecular design strategy by introducing ring-locked carbon–carbon double bonds into acceptors to enhance chemical and photochemical stability. However, these OSCs sacrificed efficiency for stability. Materials suitable for OSC commercialization must be inexpensive and available in large volumes, such as classic donor poly(3-hexylthiophene) (P3HT),14 the only organic photovoltaic material that is commercially available at a 10 kg scale. Most innovations to achieve high efficiency of NFAs also increase the synthesis complexity (SC) and cost, resulting in the preparation of NFAs only at the milligram scale and typically taking more than a week. Designing new NFAs with low cost or low SC has drawn great attention. Li et al.37 reported two low-cost acceptors MO-IDIC and MO-IDIC-2F that were synthesized by simplifying the synthetic route towards the central core. NFAs with partially or fully unfused backbones have recently been explored due to their simple molecular structures.38–41 For example, Lu et al.39 developed a few non-fused-ring electron acceptors synthesized from single aromatic units. However, large-scale preparation of NFAs that meets both low cost and low SC has not been realized. Recently, our group proposed the “all-fused-ring electron acceptor (AFAR)” concept and reported an AFRA (ITYM) (ITYM = 2,2′-(7,7,15,15-tetrahexyl-7,15-dihydro-s-indaceno[1,2-b:5,6-b′]diindeno[1,2-d]thiophene-2,10(2H)-diylidene)dimalononitrile)42–44 displaying intrinsically better chemical, photochemical, and thermal stability than those based on INCN-type terminals, and a promising PCE close to 10%. Here, we report the design of two new AFRAs named F11 and F13 by introducing a benzothiadiazole-based core (Figure 1)45 to achieve near-infrared responsiveness for promoting PCE. To tackle the issue of the large-scale preparation for NFAs, we developed an entire preparation route that enables the scalable preparation of F11 and F13 at a 10 g scale in the lab within a notably short period, demonstrating the lowest cost among reported NFAs (even comparable to that of P3HT) as well as smaller SC values than those of other low-cost NFAs.37,38 The higher highest occupied molecular orbital (HOMO), the lower optical gap with the main absorption located in the near-infrared region, and compact 3D honeycomb-type stacking with exclusively favorable J-aggregation allowed the PCEs of F13 to exceed 13%. Inspiringly, this is the efficiency record among non-INCN acceptors including fullerene-,46–48 perylene diimide (PDI)-,49–51 and rhodanine-type52–55 acceptors that also play a critical role in the development of OSCs. Compared with ITYM, F11 and F13 exhibit compact stacking and excellent photochemical stability in thin films, which contributes to the superior device stability as evidenced by miniscule burn-in losses. Figure 1 | Design of dodecacyclic AFRAs, F11 and F13. Download figure Download PowerPoint Experimental Methods 4,7-Dibromo-5,6-dinitrobenzo[c][1,2,5]thiadiazole (yield: 74%) and tributyl(thieno[3,2-b]thiophen-2-yl)stannane were synthesized by reported methods.41 Unless stated otherwise, starting materials were obtained from Anhui Zesheng Technology Co., Ltd., Shanghai Sinopharm Chemical Reagent Co., Ltd., Beijing J&K Scientific Ltd., and so on and were used without further purification. Anhydrous tetrahydrofuran (THF) and toluene were distilled over Na/benzophenone prior to use. Anhydrous dimethylformamide (DMF) was purchased from Energy Chemical. Synthesis of compound 1 4,7-Dibromo-5,6-dinitrobenzo[c][1,2,5]thiadiazole (10.2 g, 26.6 mmol) and tributyl(thieno[3,2-b]thiophen-2-yl)stannane (56.5 mmol) were dissolved in toluene (60 mL) in nitrogen atmosphere. Then tris(dibenzylideneacetone)dipalladium (469 mg, 0.5 mmol) and tris(2-methylphenyl)phosphine (622 mg, 2.0 mmol) were added to the above mixture. The reaction solution was heated at 90 °C for 15 min, and then toluene was removed under reduced pressure. The residue was filtered by methane and washed with dichloromethane to obtain 10.7 g black crude product. Synthesis of compound 2 Compound 1 (10.7 g, 21.3 mmol) and triphenylphosphine (44.5 g, 170 mmol) were dissolved in 1,2-dichlorobenzene (70 mL). The reaction mixture was purged with nitrogen for 2 min and then was heated at 180 °C for 2 h. Then reddish-brown powder (6.7 g) was obtained by immediate thermal filtration and washing with dichloromethane. Synthesis of compound 3 The crude product 2 (6.7 g, 15.3 mmol), potassium carbonate (31.2 g, 226.1 mmol), potassium iodide (125 mg, 0.75 mmol) and 1-bromo-2-hexyldecane (14 g, 45.9 mmol) in DMF (50 mL) solvent was heated to 140 °C for 2 h. The reaction solution was extracted with ethyl acetate. The organic layer was washed with water and brine and was dried over MgSO4. The solvent and excess 1-bromo-2-hexyldecane was removed by distillation under reduced pressure. The resulting mixture in dichloromethane was filtered through a diatomite layer to provide almost pure orange liquid (12.4 g). 1H NMR (400 MHz, CDCl3): δ 7.42 (m, 4H), 4.63 (d, 3J = 7.6 Hz, 4H), 2.06 (m, 2H), 1.18–0.68 (m, 60H); 13C NMR (100 MHz, CDCl3): δ 147.7, 141.7, 136.7, 131.9, 124.7, 124.3, 123.7, 121.4, 111.6, 54.9, 38.7, 31.8, 31.6, 30.4, 29.7, 29.4, 29.1, 25.5, 25.5, 22.6, 22.5, 14.1, 14.0. Synthesis of compound 4a 2-Bromobenzoyl chloride (7.74 g, 35.2 mmol) and aluminum chloride (9.3 g, 70.5 mmol) were dissolved in dichloromethane (70 mL). Then compound 3 (12.4 g, 14.0 mmol) in dichloromethane was added dropwise to the above solution over 10 min. Immediately, the solution was quenched with ice water. The organic layer was washed with water and brine, and was dried over MgSO4. The organic layer was filtered through a diatomite layer and then the product was solidified by filtration with methanol (16.6 g, 50% over four steps). 1H NMR (400 MHz, CDCl3): δ 7.72 (d, 3J = 8.0 Hz, 2H), 7.68 (s, 2H), 7.53 (dd, 3J = 7.2 Hz, 4J = 2.0 Hz, 2H), 7.48 (t, 3J = 7.2 Hz, 2H), 7.41 (t, 3J = 7.6 Hz, 2H), 4.67 (d, 3J = 8.0 Hz, 4H), 2.07 (m, 2H), 1.18–0.68 (m, 60H); 13C NMR (100 MHz, CDCl3): δ 187.8, 147.5, 142.6, 141.6, 140.2, 136.8, 133.6, 133.1, 131.5, 131.5, 130.5, 128.9, 128.4, 127.3, 119.7, 112.6, 55.3, 38.9, 31.7, 31.5, 30.4, 29.7, 29.3, 29.3, 29.12, 25.5, 25.4, 22.6, 22.5, 14.1, 13.9. Synthesis of compound 4b The synthetic route of 4b was similar to that of compound 4a (51% for four steps). 1H NMR (400 MHz, CDCl3): δ 7.69 (s, 2H), 7.55 (dd, 3J = 8.4 Hz, 4J = 5.6 Hz, 2H), 7.48 (dd, 3J = 8.4 Hz, 4J = 2.4 Hz, 2H), 7.20 (td, 3J = 8.0 Hz, 4J = 2.4 Hz, 2H), 4.67 (d, 3J = 8.0 Hz, 4H), 2.06 (m, 2H), 1.18–0.68 (m, 60H); 13C NMR (100 MHz, CDCl3): δ 186.8, 164.5, 161.9, 147.5, 142.4, 141.7, 136.7, 136.4, 136.4, 133.1, 131.6, 130.5, 130.5, 130.4, 128.6, 121.2, 121.0, 120.8, 120.7, 114.8, 114.6, 112.6, 55.3, 39.0, 31.7, 31.5, 30.4, 29.7, 29.3, 29.3, 29.1, 25.5, 25.4, 22.6, 22.5, 14.1, 14.0. Synthesis of compound F11 To a solution of compound 4a (16.6 g, 13.2 mmol) in N,N-dimethylacetamide (50 mL) was added palladium(II) acetate (286 mg, 1.3 mmol), tricyclohexylphosphonium tetrafluoroborate (938 mg, 2.5 mmol), and potassium carbonate (5 g, 36.2 mmol) in nitrogen atmosphere and then was heated at 180 °C for 15 min. Then the mixture was poured into water and filtered, and the solid was washed with water and methanol and was directly used for the next step. Pyridine (4 mL) and titanium tetrachloride (6 mL) were added to a mixture of crude product and malononitrile (3.9 g, 59.4 mmol) in chlorobenzene (50 mL). Within 5 min, the solution was extracted with dichloromethane (80 mL) and water. The solvent was removed under reduced pressure. The residue in dichloromethane solvent went through a fast filtration with diatomite and then was purified with recrystallization (petroleum ether/dichloromethane, 4:1 v/v) to give the target molecule (10.4 g, 66%). 1H NMR (400 MHz, CDCl3): δ 8.19 (d, 3J = 7.6 Hz,, 2H), 7.46 (t, 3J = 7.6 Hz, 2H), 7.38 (d, 3J = 7.2 Hz, 2H), 7.31 (t, 3J = 7.6 Hz, 2H), 4.65 (d, 3J = 7.6 Hz, 4H), 2.09 (m, 2H), 1.18–0.68 (m, 60H); 13C NMR (100 MHz, CDCl3): δ 157.1, 147.3, 147.2, 138.3, 137.3, 136.5, 136.4, 134.6, 134.2, 133.6, 133.2, 129.6, 128.9, 126.3, 121.2, 114.2, 113.4, 113.3, 70.9, 55.7, 39.1, 31.8, 31.5, 30.6, 30.5, 30.4, 29.7, 29.4, 29.2, 25.7, 25.6, 25.5, 25.5, 22.6, 22.5, 14.0, 13.9; high-resolution mass spectrometry (HRMS) [matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF)] (m/z): [M]+ calcd for C70H74N8S5, 1186.4634; found, 1186.4627. Anal. Calcd for C70H74N8S5 (%): C, 70.79; H, 6.28; N, 9.43. Found: C,70.48; H, 6.24; N, 9.19. Synthesis of compound F13 The synthetic processes was similar to that of F11 (61%). 1H NMR (400 MHz, CDCl3): δ 8.19 (dd, 3J = 8.4 Hz, 4J = 4.4 Hz, 2H), 7.13 (dd, 3J = 7.6 Hz, 4J = 2.4 Hz, 2H), 6.97 (td, 3J = 8.4 Hz, 4J = 2.4 Hz, 2H), 4.66 (d, 3J = 8.0 Hz, 4H), 2.07 (m, 2H), 1.18–0.68 (m, 60H); 13C NMR (100 MHz, CDCl3): δ 167.2, 164.6, 156.0, 147.3, 145.1, 139.2, 139.1, 137.2, 136.4, 135.7, 134.2, 133.9, 133.8, 133.6, 129.5, 128.0, 127.9, 114.5, 114.3, 114.0, 113.4, 113.1, 110.3, 110.1, 71.3, 55.8, 39.2, 31.8, 31.5, 30.6, 30.6, 30.5, 30.4, 29.7, 29.3, 29.3, 29.2, 25.7, 25.6, 25.5, 25.5, 22.6, 22.5, 14.0, 13.9; HRMS (MALDI-TOF) (m/z): [M]+ calcd for C70H72F2N8S5, 1222.4446; found, 1222.4445. Anal. Calcd for C70H72F2N8S5 (%): C, 68.71; H, 5.93; N, 9.16. Found: C, 68.70; H, 5.91; N, 8.86. F13 shows a good solubility in various commonly used solvents such as chloroform (38 mg/mL), THF (25 mg/mL), toluene (20 mg/mL), chlorobenzene (20 mg/mL), and dichlorobenzene (30 mg/mL). Results and Discussion Large-scale preparation and single-crystal X-ray analysis of AFARs F11 and F13 Because large-scale preparation requires optimization of each synthetic procedure, not just a certain step, we optimized the synthetic route for F11/F13 and the detailed processes are shown in Figure 2. In the first step, we selected an efficient catalytic system of tris(benzylideneacetone)dipalladium and tris(2-methylphenyl)phosphine for the Stille-type cross-coupling reaction, in which the starting materials of 4,7-dibromo-5,6-dinitrobenzo[c][1,2,5]thiadiazole and tributyl(thieno[3,2-b]thiophen-2-yl)stannane are accessible commercially or can be prepared in the lab on a large scale. We took advantage of the solubility difference between the reactants and the product to obtain compound 1 by direct filtration. In comparison with the corresponding synthesis of Y6, the reaction time was drastically reduced from 12 h to 15 min. In the second step of nitrocyclization, we optimized the reaction by replacing triethyl phosphate with the readily available triphenylphosphine. The o-dichlorobenzene solution of compound 1 and triphenylphosphine was heated at 180 °C for 2 h. High temperature affects the product purification equivalent to recrystallization, and the reddish-brown powder of 2 was directly obtained by immediate filtration. In the third step of alkylation, the extracted organic layer was passed through thin-layer diatomite to remove a small amount of excess potassium carbonate. Because this reaction is clean and shows no by-products, the resulting product 3 was directly used for the next step without further purification. For the fourth step of Friedel–Crafts acylation, a dichloromethane solution of compound 3 was added dropwise to that of aluminum trichloride and 2-bromobenzoyl chloride. Immediately, a crude product of 4a was solidified by filtering and washed with methanol due to a high reaction yield and purity, thereby ensuring that further purification was not necessary. A total yield of 50% for the presented four steps was achieved. The fifth step involves successive intramolecular cyclization and Knoevenagel condensation. We selected Pd-catalyzed C–H activation for intramolecular cyclization in consideration of its great potential for a green, sustainable, and atom-efficient synthesis at large scale.56 After reacting for 15 min, the crude product was directly used for the next condensation without any purification as a result of the efficient nature of C–H-activated cyclization. A different reaction than the condensation occurred in INCN-type NFAs, and this condensation was completed in an extremely short time (<5 min) at room temperature. Finally, the target molecule F11 was purified by recrystallization in dichloromethane/petroleum ether after passing through diatomite. It is noteworthy that the whole synthetic process eliminates the need for column chromatography, which is perfectly manageable at the laboratory scale yet very expensive or even impossible at the industrial scale.57 Finally, all these synthetic advantages throughout the whole procedure helped achieve the goal of the large-scale preparation of acceptors, in which over 10 g of F11 was obtained in the lab. A total yield of 33% was produced despite a certain waste of products caused by recrystallization, equivalent to >80% yield for each step. Moreover, the synthetic process takes an extremely short time for the rapid preparation of NFAs at a large scale, while most NFAs take more than one week.7 The synthetic process of F13 was similar to that of F11, and F11 and F13 were fully characterized by using 1H NMR, 13C NMR, matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF-MS) and elemental analysis, as shown in Supporting Information Figures S21–S30. We believe that such a facile and efficient synthetic process can also be transferred to kg-scale industrial production. Figure 2 | Depiction of quick and macroscopic preparation of the all-fused-ring acceptors, F11 and F13. Reagents and conditions: (a) Pd2(dba)3, P(o-tol)3, tributyl(thieno[3,2-b]thiophen-2-yl)stannane, toluene; (b) PPh3, o-dichlorobenzene (c) HD-Br, K2CO3, N,N-dimethylformamide, HD = 2-hexyldecanyl; (d) AlCl3, o-bromoaryl chloride, dichloromethane; (e) (1) K2CO3, N,N-dimethylacetamide, Pd(OAc)2, PH(cychex)3BF4; (2) Malononitrile, TiCl4, pyridine, chlorobenzene. Download figure Download PowerPoint We examined the single-crystal structure of F11 by X-ray diffraction. The single-molecule configurations of F11, without alkyl chain for clarity, are shown in Figure 3a, displaying the extra-long dodecacyclic structure. In contrast to the reported Y-type acceptors, the configuration of all-fused-ring F11 presents a W-shaped and planar structure. Figure 3b exhibits the two types of J-aggregations between the end groups of adjacent molecules, with the π–π stacking distances of 3.39 and 3.30 Å, respectively. In addition, there are intermolecular N···N (3.48 and 3.59 Å) and S···S (3.56 Å) interactions. As a result, a regular hexagonal frame with three molecular layers in the perpendicular direction is induced by the J-aggregation effect and intermolecular interactions, as shown in Figure 3c. In comparison with the one-dimensional (1D) stacking of ITYM, the 3D network packing structure of F11 shown in Figure 3d, which looks like a honeycomb, is beneficial for charge transport in multiple channels. Figure 3 | (a) The single-molecule configurations of F11, without alkyl chain for clarity (CCDC number: 2120638). (b) The intermolecular interactions between F11 molecules. (c) The single-crystal structure of one regular hexagonal frame from different views. (d) The 3D network packing structure of F11. Download figure Download PowerPoint Cost analysis We analyzed the costs of F11 and F13 based on macroscopic preparation. Regarding only the cost of reagents and solvents used in the whole synthetic procedure, we calculated the lab-scale total costs Cg ($/g) of F11 and F13 according to actual consumption, and the details are listed in Supporting Information Table S1. Compared with the Cgs reported for other organic photovoltaic materials,37 F11 and F13 have incredibly low costs, with a Cg of 11.4 $/g for F11 and 11.5 $/g for F13, which are much lower than other NFAs and are almost at the same level as P3HT (10.8 $/g), as illustrated in Figure 4a and Supporting Information Table S2. It should be noted that these cost values will be further reduced if performed on an industrial mass production basis. The extremely low costs of our acceptors are partly because of the simplification of the purification processes, especially the absence of column chromatography. According to the statistics,37 the costs of purification with column chromatography in most NFAs account for over 30%, even up to 50% of the total costs, as a result of consuming large amounts of silica gel and solvents. In contrast, the design and selection of o-bromoaryl chloride as terminal units is also an important reason for the cost reduction. From the material suppliers, such as Aladdin ( Supporting Information Table S3), o-bromoaryl chloride costs about 3 $/g based on a 5-g package, which is quite economical compared with INCN (138 $/g). It should be noted that labor costs that usually are ignored by the academic community account for a large percentage of the total cost in actual industrial production. The scalable preparation of F11 and F13 was finished within extremely short time, thus significantly reducing labor costs. In addition, we assessed the synthetic complexity (SC) of our F-acceptors along with Y6, ITIC, D18, and PTQ10 (more detailed information is shown in Supporting Information Figures S1–S9 and Tables S4 and S5). Both F11 and F13 show notably lower SC values (29.7% for F11 and 33.4% for F13) than that of Y6 (82.4%) and even lower SC values than other low-cost NFAs, which is consistent with the low costs of F-acceptors. Figure 4 | (a) Histogram of material costs Cg ($/g) for the different photovoltaic materials (material costs are adopted from the literature,37 except for our acceptors). (b) Statistics of high PCEs for typical non-INCN series photovoltaic acceptors ( Supporting Information Table S6). Download figure Download PowerPoint Photophysical, electrochemical, and molecular stacking properties Figure 5a shows the normalized absorptions of F11, F13, and ITYM in film. The maximum absorption peaks of F11 and F13 are 773 nm, which are significantly redshifted by 93 nm in comparison to that of ITYM. The optical bandgaps are both 1.44 eV for F11 and F13 with the main absorption located in the near-infrared region, which is much smaller than 1.68 eV for ITYM. The frontier orbital energy levels of F11 and F13 were calculated by cyclic voltammetry ( Supporting Information Figure S10 and Figure 5b). In comparison with ITYM, the elevated HOMO of F11 and F13 are −5.64 and −5.65 eV, respectively, and the lowest unoccupied molecular orbital (LUMO) energy levels are −3.78 and −3.81 eV for F11 and F13, respectively. We ascribed the deeper energy levels of F13 than F11, especially the LUMO, to the electron-withdrawing ability of the F atom. Figure 5 | (a) Normalized UV–vis–NIR absorption spectra. (b) Energy diagram of D18, ITYM, F11, and F13. (c) 2D GIWAXS patterns and (d) GIWAXS intensity profiles along the in-plane and out-of-plane directions of pristine films. Download figure Download PowerPoint The molecular stacking behaviors of F11, F13, ITYM, and D18 in film were measured by grazing incidence wide-angle X-ray scattering (GIWAXS), and GIWAXS patterns of D18 are shown in Supporting Information Figure S17. Figure 5c shows 2D GIWAXS images and Figure 5d shows 1D profiles of the in-plane (IP) and out-of-plane (OOP) direction of different neat films. It is obvious that ITYM in film shows a poor diffraction peak, indicating loose molecular stacking and weak crystallinity. While the neat films of F11 and F13 exhibit a favorable face-on orientation with strong (010) diffraction peaks in the OOP direction at 1.74 and 1.79 Å−1, respectively. F13 film displayed a larger crystal coherence length (CCL = 27.4 Å) calculated from the Scherrer equation and the smaller d-spacing (d = 3.51 Å) for (010) diffractions than those of F11 (CCL = 24.3 Å and d = 3.61 Å), meaning a higher crystallinity of the F13 film, which might be induced by the F atom. Chemical, photochemical, and thermal stability Our previous work confirms that avoiding highly polarized and isolated double bonds facilitates the stability of the material. Herein, the chemical and photochemical stability of F11 and F13 were examined, and two typical acceptors based on the INCN-group, ITIC and Y6, were chosen for comparison. Chemical stability measurements were performed by monitoring the absorption spectra of four acceptors before and after treatment with ethanolamine (EA) in the corresponding THF solutions are shown in Supporting Information Figure S11. Figure 6a depicts the time-dependent absorption decays of acceptors at the corresponding maximum absorptions. Immediately after the addition of EA (100 equiv), a significant color change ( Supporting Information Figure S13a) and decay of absorption occurred in Y6 and ITIC solutions. Almost no attenuation of the maximum absorption peak intensity of F11 and F13 was found, even 12 h after adding EA. However, obviously decreased intensities of 83% and >99% for Y6 and ITIC, respectively, were observed. Furthermore, the absorption at 350–500 nm increased to a different extent, which can be attributed to interruption of the D–A conjugation. Figure 6 | The time-dependent absorption decays of acceptors at the corresponding maximum absorptions upon the (a) EA treatment in THF:H2O mixtures (96:4, v/v) (the concentration of NFAs is controlled at 10−5 M, while that of EA is 10−3 M) and (b) 300 mW cm−2 (LED) irradiation in films. Download figure Download PowerPoint Photostability of photovoltaic-active materials under ambient conditions plays an important role in OSCs. Thus, we investigated the photochemical stability of the four acceptors (F11, F13, ITIC, and Y6) for comparison, and Supporting Information Figure S12 displays the evolution of absorption spectra in THF solutions and in films with irradiation time. After irradiating for 10 min under AM 1.5G at 100 mW cm−2, the absorption peak of air-saturated ITIC solution in the 600–700 nm range disappeared, suggesting the D–A conjugation was broken. The maxi