English

• Organic solar cells (OSCs) have attracted widespread attention due to their advantages, such as low cost, light weight, solution processing, and large-area printing [1–6]. In the past few years, active layer materials have blossomed. A series of excellent polymer donors represented by PM6, PTQ10, and D18 have been developed, and multitudinous non-fullerene acceptors (NFAs) have sprung up [7–9]. The currently popular NFAs mainly apply A (acceptor)-D (donor)-A (acceptor) type and A-DA’D-A type skeleton structures [10]. The strong intramolecular charge transfer (ICT) endows these NFAs with broad absorption spectra, facilitating light-harvesting and thus significantly enhancing the short-circuit current (J_SC) [11]. So far, the power conversion efficiency (PCE) of single-junction and tandem OSCs based on A-DA’D-A type acceptors has exceeded 19% and 20%, respectively, under the standard AM 1.5G illumination [12–14]. Although the energy loss of the corresponding organic photovoltaic system has been reduced to around 0.5 eV, the open-circuit voltage (V_OC) is always below 1.0 V [15]. Once these OSCs work under indoor light conditions, V_OC will further decrease [16,17]. For example, PM6:Y6-O-based devices under 1-Sun condition can achieve a PCE of 16.53% with a V_OC of 0.94 V. Under 1650 lux 3000 K LED illumination, the PM6:Y6-O-based indoor device achieved a high PCE of 30.89%, but the V_OC is down to 0.84 V [18]. The ultra-low V_OC will increase the operating cost of industrialization. In addition, because the spectra of indoor light sources are located in the 400–750 nm range, such narrow bandgap acceptor materials could incur more losses due to spectral redundancy [10]. Hence, it is necessary to develop high-voltage wide-bandgap (WBG) photovoltaic materials that meet the requirements of indoor photovoltaics, which can fabricate high-efficiency indoor photovoltaics (IPVs) to power microelectronic devices in the Internet of Things (IoTs) [19,20].

  Recently, several kinds of WBG NFAs with high lowest unoccupied molecular orbital (LUMO) energy levels have been developed through the precise design of electron-donating and electron-withdrawing units. For example, our group chose weak electron-withdrawing benzotriazole (BTA) as the A₁ unit and rhodanine derivatives as the A₂ unit to couple with indacenodithiophene (IDT) electron-donating unit and achieved some WBG A₂-A₁-D-A₁-A₂ type NFAs [21–23]. Due to the slightly weak intramolecular electron push-pull effect, these acceptors typically have wider bandgaps (~1.7–1.8 eV) and higher LUMO levels (LUMO > −3.8 eV). When blended with the donors having deep HOMO levels, the corresponding photovoltaic devices exhibited high V_OC exceeding 1.1 V. In 2019, we achieved a high V_OC of 1.24 V and a PCE of 10.5% by using the same A strategy to mix J52-Cl with BTA3 [24]. In 2023, BTA3–4F was synthesized by introducing F atoms into the side chains on the IDT unit. The photovoltaic device prepared with the terpolymer E18 and BTA3–4F obtained a V_OC of 1.3 V with a PCE of 10%, a breakthrough in this field [25]. These BTA-based wide-bandgap NFAs can be used as a third component to prepare ternary photovoltaic devices, effectively improving V_OC and assisting PCE breakthrough by over 19% [26]. Importantly, these BTA-based NFAs exhibited excellent performance in indoor photovoltaics. In 2021, PBDB-T:BTA3 achieved a PCE of over 25% under 1000 lux 2700 K LED illumination, and the system can still maintain a V_OC of about 1 V [27]. In the same year, Hou and his group designed a new wide-bandgap polymer PBQx-TCl to mix with BTA3, and the 1 cm² device obtained a 28.5% PCE with a V_OC of 1.10 V under indoor lighting [28]. In 2022, the tetrahydrofuran-processed ternary OSCs based on J52-Cl:BTA3:BTA1 afforded a PCE of 28.8% with a higher V_OC of 1.12 V under irradiation of a 1000 Lux light-emitting diode (LED) light [29]. Besides, D18: Cl-BTA5 devices can achieve a V_OC of 1 V and a PCE of 21% under indoor lighting conditions [20]. The large-area device (1 cm²) that combined low-cost polymer PTQ10 with non-fused NFAs acceptor Cl-BTA33 can achieve a PCE of over 24% under indoor light exposure. Therefore, further exploring the structure-property relationship of these BTA-based molecules can help design new promising materials and develop high-efficiency IPVs.

  As an effective molecular design strategy, halogenation has been widely applied in material design. The variation of positions and quantities of halogen atoms can lead to diverse photophysical properties and energy losses, leading to quite different device performance, for example, the classical NFAs, ITIC vs. IT-4F and Y5 vs. Y6 [30–32]. Our previous works have investigated the effect of halogenated on the BTA unit and central IDT unit in the BTA-based A₂-A₁-D-A₁-A₂ type NFAs [22]. The data indicated that introducing halogen atoms into the BTA unit had little influence on bandgap but lowered both the highest occupied molecular orbitals (HOMO) and LUMO energy levels, which can increase the charge transfer driving force and non-radiative recombination energy loss [33]. Introducing F atoms into the side chains on the central nucleus can increase molecular bandgap and change intermolecular interactions and stacking forms. Ultimately, both can improve device performance [34–36]. BTA5, with the benzyl group-substituted dicyanomethylene rhodanine as terminals, has shown improved device performance than BTA3 because the benzyl group could act on the intermolecular interaction patterns and facilitate the charge generation [37]. Further modifying the benzyl side chain and lucubrating their structure-performance relationship may achieve high photovoltaic performance.

  Herein, with BTA5 as a benchmark, we study the impact of the chlorine atoms quantities in the terminal benzyl groups on device performance. Two BTA-based acceptor molecules, as shown in Fig. 1a, BTA5-Cl containing one chlorine atom in the benzyl groups and BTA5-2Cl bearing the two chlorine atoms in the benzyl groups, are synthesized. We chose D18-Cl (Fig. 1b) as the donor material to pair with the above three NFAs due to its deep HOMO level, high hole mobility, and strong crystallinity. D18-Cl:BTA5 has afforded the highest V_OC of 1.28 V with the smallest non-radiative recombination loss (ΔE₃) of 0.14 eV. ΔE₃ slightly increased from 0.14 eV to 0.17 and 0.19 eV, respectively, with the increase of chlorine atoms. As a result, the V_OC of D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl are decreased to 1.22 V and 1.20 V, respectively. As the number of chlorine atoms in the terminals ascends, the molecular energy levels shift downward gradually, and the charge transfer driving force increases in order. At the same time, carrier mobilities become more balanced. Thus, the J_SC and fill factor (FF) increase sequentially. Finally, the D18-Cl:BTA5-2Cl device achieves the highest PCE of ~11%, with a V_OC of 1.20 V. These data indicate that the terminal side chains chlorination also plays an important role in the device's performance. Due to the matched absorption with the indoor sources, these high-V_OC OSCs have good potential for indoor OSC application.

  Figure 1

  

  Figure 1.  Chemical structures of BTA5, BTA5-Cl and BTA5-2Cl (a) and the polymer D18-Cl (b). The absorption spectra of donor and three acceptors in chloroform solution (c) and neat films (d). (e) Energy levels diagram of D18-Cl, BTA5, BTA5-Cl and BTA5-2Cl.

  DownLoad: Full-Size Img PowerPoint

  The synthesis routes of the intermediate and small molecule acceptor BTA5-2Cl are provided in Fig. S1 (Supporting information). The ¹H NMR spectrum of the intermediate is depicted in Fig. S2 (Supporting information). BTA5 and BTA5-Cl were reported in our previous works [38]. The ¹H NMR, ¹³C NMR, and mass spectra of BTA5-2Cl are depicted in Figs. S3-S5 (Supporting information). Density functional theory (DFT) calculation based on the B3LYP/6–31G(d,p) basis set is performed on BTA5-2Cl. The calculated results in Fig. S6a (Supporting information) indicate that introducing different numbers of chlorine atoms on the end groups does not significantly alter the molecular conformation, and the three acceptor materials still exhibit a planar skeleton with negligible steric hindrance effects. Due to the increase in the number of halogen atoms, the electron-accepting ability of the corresponding acceptor is enhanced, and the energy levels of BTA5, BTA5-Cl, and BTA5-2Cl shift downwards, which could increase the energy offsets between the donors and acceptors and thus act on the V_OC and J_SC. In addition, we mapped the molecular surface electrostatic potential (ESP) distributions of the three acceptors (Fig. S6b in Supporting information). The overall ESP values on the molecular surfaces are almost positive for the three acceptors, and the overall average ESP values on the molecular surface increase with the addition of Cl atoms due to the electron-withdrawing property of the Cl atom. Gradually increasing ESP values could enhance intermolecular interactions with the polymer donor [39].

  The ultraviolet-visible (UV–vis) absorption spectra of three systems in the CHCl₃ solution and thin film state are shown in Figs. 1c and d. The shape of the absorption spectrum has not undergone significant changes with the increase in the number of chlorine atoms in the side chain on terminals. As shown in Fig. 1c, the maximum absorption peaks of BTA5, BTA5-Cl, and BTA5-2Cl in the chloroform solution are 618, 622, and 627 nm, respectively, showing slight red-shifts as the increase of the Cl atoms. Compared with the solution absorption, the maximum absorption peaks of three acceptors in films significantly red-shift to 643, 648, and 651 nm, respectively, which indicates that all acceptors have strong intermolecular interactions in the thin films. With the increase of chlorine atoms, the absorption onsets are red-shifted. As a result, the optical bandgaps of BTA5, BTA5-Cl, and BTA5-2Cl are reduced from 1.84 eV to 1.77 and 1.73 eV, respectively. In addition, it is worth noting that BTA5-2Cl has the highest molar extinction coefficient of 1.82 × 10⁵ L mol⁻¹ cm⁻¹, while BTA5 and BTA5-Cl have similar molar absorption coefficients (1.35 × 10⁵ L mol⁻¹ cm⁻¹ vs. 1.40 × 10⁵ L mol⁻¹ cm⁻¹). The higher extinction coefficient can enhance the light capture ability of the acceptor, thereby contributing to J_SC.

  The effect of chlorination on the electrochemical property is investigated with cyclic voltammetry (CV). The CV curve of BTA5-2Cl is shown in Fig. S7 (Supporting information). As shown in Fig. 1e, the HOMO level of D18-Cl is −5.49 eV, and the LUMO level is −3.49 eV [40]. The E_HOMO/LUMO energy levels are −5.55/−3.71, −5.65/−3.88, and −5.72/−3.99 eV for BTA5, BTA5-Cl [38], and BTA5-2Cl, respectively, implying that molecular energy levels decrease in order as Cl atoms increase. The variation trend of the molecular energy levels is consistent with the DFT calculation result. These data indicate that adding Cl atoms may provide a greater driving force for charge dissociation and help obtain higher J_SC but may sacrifice the V_OC.

  To further estimate the influence of the Cl atoms on the photovoltaic performance, conventional devices are prepared with the structure of indium tin oxide (ITO)/poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS)/active layers/PFN-Br/Ag. The detailed device optimization process is shown in Tables S1-S3 (Supporting information), and the optimal device preparation conditions are provided in the supporting information. The optimal photovoltaic parameters of the three systems are listed in Table 1, and the corresponding current density-voltage (J-V) curves are presented in Fig. 2a. The devices based on D18-Cl:BTA5, D18-Cl:BTA5-Cl, and D18-Cl:BTA5-2Cl blends achieved a high V_OC of 1.28, 1.22, and 1.20 V, respectively. Increasing chlorine atoms on the terminal benzyl groups of the acceptors could gradually reduce the V_OC, which is attributed to the descending LUMO energy levels and increased energy loss. The J_SC values of the three devices rise as the number of Cl atoms increases, partly related to the increase of energy offsets between the HOMO levels of D18-Cl and these acceptors. D18-Cl:BTA5-2Cl device presents the highest J_SC value (12.89 mA/cm²), higher than D18-Cl:BTA5 (7.38 mA/cm²) and D18-Cl:BTA5-Cl (11.53 mA/cm²). At the same time, the D18-Cl:BTA5-2Cl-based device achieves the highest FF of 0.71. D18-Cl:BTA5-Cl comes second (FF = 0.65), and D18-Cl:BTA5 is the worst (FF = 0.53). At last, the PCE of the D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl blends are 5.07%, 9.15% and 10.96%, respectively. Fig. 2b shows the external quantum efficiency (EQE) curves of the optimized devices. The maximal EQE values of the three devices are 40%, 60%, and 67% for D18-Cl:BTA5, D18-Cl:BTA5-Cl, and D18-Cl:BTA-2Cl, respectively. The integrated current density (J_cal) values calculated from the EQE diagram are consistent with the current values calculated from the J-V curves.

  Table 1

  

  Table 1.  Photovoltaic parameters of the optimized three OSCs.

  DownLoad: CSV

  Figure 2

  

  Figure 2.  J-V curves (a) and EQE spectra (b) of the OSCs based on D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl. (c) The PL of neat films and blend films with excitation at 650 nm. J^1/2-V curves of electron-only (d) and hole-only (e) mobilities tested by the SCLC method.

  DownLoad: Full-Size Img PowerPoint

  Steady-state photoluminescence (PL) spectroscopy testing is used to investigate whether exciton dissociation had occurred in three mixed films. As shown in Fig. 2c, when selectively excited the acceptor with the excitation wavelength of 650 nm, the PL quenching rates of the three mixed films (D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl) are calculated to be 62%, 86%, and 91%, respectively, indicating the gradually improved hole transfer efficiency. Increasing the number of Cl atoms on the terminals helps generate a more effective hole transfer, thereby facilitating the photocurrent generation.

  We further apply the space-charge-charge-limited-current (SCLC) method to calculate the electron and hole mobility of three mixed films and study the effect of the number of chlorine atoms on the charge transport property [41]. The detailed device preparation process and calculation method are summarized in the supporting information. As shown in Figs. 2d and e, the electron mobilities (µ) of D18-Cl:BTA5, D18-Cl:BTA5-Cl, and D18-Cl:BTA5-2Cl are 1.79 × 10⁻⁴, 2.94 × 10⁻⁴, and 3.85 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively, and the corresponding hole mobilities are 3.06 × 10⁻⁴, 4.23 × 10⁻⁴, and 4.98 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. The results show that with the increase of chlorine atoms in the acceptors, the electron and hole mobilities of the corresponding devices increase, and the electron mobilities increase more significantly. Thus, the ratio of µ/µ (0.58, 0.7, and 0.77) approaches closer and closer to 1 as the number of chlorine atoms increases. The more balanced carrier mobility in D18-Cl:BTA5-2Cl helps reduce charge recombination and facilitates higher J_SC and FF.

  In order to understand the impact of the number of chlorine atoms on the charge recombination process, we studied the relationship between J_SC or V_OC and the light intensity (P_light). The relationship between J_SC and P_light can be analyzed using the formula (Eq. 1):

  (1)

  When the value of α is closer to 1, bimolecular recombination is nearly inhibited [42]. As shown in Fig. S8 (Supporting information), the α values are 0.954, 0.966, and 0.967 for the devices based on D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl, respectively. The D18-Cl:BTA5-2Cl device, with the highest ɑ value, exhibits the lowest bimolecular recombination. The V_OC and light intensity relationship can be expressed with the formula (Eq. 2):

  (2)

  The larger the n value, the more severe the monomolecular recombination or trap capture in the system, indicating a lower conversion efficiency of the device [43]. The n values of the three systems can be calculated as 1.45, 1.36, and 1.24, respectively. Compared with the other two systems, the monomolecular and trap-assisted recombination in D18:BTA5-2Cl is significantly suppressed, improving the J_SC and FF.

  In order to investigate the effect of the number of chlorine atoms in the acceptor on energy loss, the FTPS-EQE spectra and EQE_EL curves were tested and depicted in Fig. S9 (Supporting information). The energy loss data of relevant devices are shown in Table S4 and Fig. S10 (Supporting information). The total energy loss can be divided into three parts (Eq. 3) [44]:

  (3)

  ΔE₁ is caused by radiative recombination loss above the bandgap, which is inevitable. It can be expressed using a formula (Eq. 4):

  (4)

  where E_g is the bandgap of the D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl, q is the element charge, and V_OC^SQ is the Shockley-Queisser (SQ) ultimate output voltage. The ΔE₁ of the related devices are 0.288, 0.285, and 0.285 eV. ΔE₂ is the radiative recombination below the bandgap, mainly caused by the absorption of the CT state. ΔE₂ can be calculated using the formula (Eq. 5):

  (5)

  where V_OC^rad is the limited V_OC under only radiation recombination conditions. ΔE₂ of the D18-Cl:BTA5, D18-Cl:BTA5-Cl, and D18-Cl:BTA5-2Cl are 0.093, 0.082 and 0.083 eV, respectively. ΔE₂ is also related to energy disorder, which can be reflected with Urbach energy (E_U). As shown in Fig. 6a, according to the formula (Eq. 6) [45]:

  (6)

  The E_U values for BTA5-, BTA5-Cl-, and BTA5-2Cl-based devices are 37.2, 36.7, and 36.8 meV, respectively. The fluctuation of E_U values is consistent with the change of ΔE₂, indicating that introducing chlorine atoms into the terminals can slightly reduce energy disorder. ΔE₃ is the non-radiative energy loss, which can be calculated from the EQE_EL spectrum, according to the following formula (Eq. 7):

  (7)

  From Fig. S10a, the EQE_EL values for BTA5-, BTA5-Cl-, and BTA5-2Cl-based devices are 4.59 × 10⁻³, 1.52 × 10⁻³, and 6.99 × 10⁻⁴, respectively. The corresponding ΔE₃ values are 0.139, 0.168, and 0.188 eV, respectively. ΔE₃ = 0.139 eV, one of the lowest values in OSCs, is roughly the same as those of inorganic or perovskite solar cells [46–50]. Introducing chlorine atoms into the terminals can slightly reduce the luminescence efficiency, increasing ΔE₃ [51]. In summary, introducing Cl atoms could gradually increase energy loss from 0.520 eV to 0.535 and 0.556 eV. Hence, BTA5-2Cl with di-chlorination attains the lowest V_OC of 1.20 V, followed by BTA5-Cl with mono-chlorination (V_OC = 1.22 V), and BTA5 without Cl atoms exhibits the highest V_OC of 1.28 V.

  Contact angle testing is commonly used to study the miscibility between donor polymers and small molecule acceptors. The contact angles of D18-Cl, BTA5, BTA5-Cl, and BTA5-2Cl pure films were tested using water and diiodomethane, as shown in Fig. S11 (Supporting information). The detailed data are listed in Table S5 (Supporting information). The test results were analyzed using the Flory-Huggins theory, according to the formula (Eq. 8):

  (8)

  where χ is Flory-Huggins interaction, K is the constant, and γ is the surface tension of the relevant material [52]. The calculated χ values of the three devices are 0.69 K for D18-Cl:BTA5, 0.56 K for D18-Cl:BTA5-Cl, and 0.59 K for D18-Cl:BTA5-2Cl, respectively. The D18-Cl:BTA5-based device shows the largest χ values among the three devices, indicating that the miscibility of the device is the worst, which may form the large phase separation. The χ value decreases with the introduction of chlorine atoms, indicating the miscibility is improved, which is beneficial for charge transport and exciton dissociation, achieving higher FF and better PCE.

  Atomic force microscopy (AFM) is used to study the surface morphology of the blend film. As shown in Figs. 3a–c, the root-mean-square (RMS) roughness of D18-Cl:BTA5, D18-Cl:BTA5-Cl, and D18-Cl:BTA5-2Cl are 1.64, 1.29, and 1.28 nm, respectively. Compared with the D18-Cl:BTA5 blend film, with the introduction of chlorine atoms, the active layer obtained a smoother surface, which is conducive to forming more effective charge extraction. These data are consistent with the data of the contact angle testing. To further investigate the influence of the number of chlorine atoms on molecular stacking and crystallinity, we conducted two-dimensional grazing-incident wide-angle X-ray scattering (2D GIWAXS) testing on three blend films. As shown in Figs. 3d–g, three blend films mainly exhibit an obvious (100) diffraction peak in the in-plane (IP) direction and a (010) diffraction peak in the out-of-plane direction, implying that all of the three films preferred the face-on orientation, which is conducive to charge transport. The detailed data is presented in Table S6 (Supporting information). BTA5, BTA5-Cl, and BTA5-2Cl exhibit similar π-π stacking distances of 3.57, 3.58, and 3.57 Å, respectively. The crystallite coherence lengths (CCL) of the (010) peak for BTA5, BTA5-Cl, and BTA5-2Cl are 20.86, 22.04, and 25.84 Å, respectively. Meanwhile, the corresponding CCLs of the (100) peak are 96.70, 104.11, and 115.42 Å in the IP direction. These data indicate that as the number of chlorine atoms increases, the crystallinity of the mixed film gradually increases, which is beneficial to charge transport, obtaining higher carrier mobilities and current values.

  Figure 3

  

  Figure 3.  Atomic force microscopy height images of D18-Cl:BTA5 (a), D18-Cl:BTA5-Cl (b) and D18-Cl:BTA5-2Cl (c). 2D GIWAXS patterns of D18-Cl:BTA5 (d), D18-Cl:BTA5-Cl (e), D18-Cl:BTA5–2Cl (f) and the corresponding 1D profiles (g).

  DownLoad: Full-Size Img PowerPoint

  In summary, we introduced different numbers of chlorine atoms on the terminal benzene rings in BTA5 and obtained two NFAs, BTA5-Cl and BTA5-2Cl. D18-Cl, with deeper HOMO energy levels and strong crystallinity, was chosen as the polymer donor to pair with these acceptors and investigate the effect of the chlorine atom's quantities on the photovoltaic performance. Due to the low energy offsets and low non-radiative energy loss below 0.20 eV, these devices attain a high V_OC above 1.20 V. As the number of chlorine atoms increases, V_OC gradually decreases from 1.28 V to 1.22 V and 1.20 V, with the non-radiative energy loss increasing from 0.14 eV to 0.17 and 0.19 eV. BTA5-2Cl-based OSCs with the most Cl atoms achieve a PCE of nearly 11% due to more effective exciton dissociation and charge transfer capabilities, as well as more balanced carrier mobility. Our research results indicate that increasing the number of halogen atoms in the acceptor is beneficial for achieving better performance while maintaining a high voltage. Despite the non-conjugated nature of the substituents, chlorination remains effective in modulating the molecular energy levels. These results could provide important guiding principles for future molecular design.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Jinge Zhu: Writing – original draft, Investigation, Data curation. Ailing Tang: Writing – review & editing, Supervision, Project administration, Funding acquisition. Leyi Tang: Methodology, Data curation. Peiqing Cong: Methodology, Investigation, Data curation. Chao Li: Methodology, Investigation. Qing Guo: Supervision, Data curation. Zongtao Wang: Methodology, Investigation, Data curation. Xiaoru Xu: Methodology, Investigation. Jiang Wu: Methodology, Data curation. Erjun Zhou: Writing – review & editing, Supervision, Project administration, Investigation, Funding acquisition.

  Acknowledgment

  The authors thank the support from the National Natural Science Foundation of China (Nos. 52373176, 52073067).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110233.

  
  
• 1. [1]

     Y. Cui, H.F. Yao, J.Q. Zhang, et al., Adv. Mater. 32 (2020) 1908205. doi: 10.1002/adma.201908205

  2. [2]

     S. Dong, K.X. Zhang, B.M. Xie, et al., Adv. Energy Mater. 9 (2018) 1802832.

  3. [3]

     G.U. Kim, J.H. Park, S. Lee, et al., J. Mater. Chem. A 10 (2022) 9408–9418. doi: 10.1039/d2ta00145d

  4. [4]

     C.Q. Li, X. Zhang, N.G. Yu, et al., Adv. Funct. Mater. 32 (2022) 2108861. doi: 10.1002/adfm.202108861

  5. [5]

     D.L. Liu, B. Yang, B. Jang, et al., Energy Environ. Sci. 10 (2017) 546–551. doi: 10.1039/C6EE03489F

  6. [6]

     L. Wang, T.T. Wang, J. Oh, et al., Chem. Eng. J. 442 (2022) 136068. doi: 10.1016/j.cej.2022.136068

  7. [7]

     J. Yuan, Y.Q. Zhang, L.Y. Zhou, et al., Joule 3 (2019) 1140–1151. doi: 10.1016/j.joule.2019.01.004

  8. [8]

     Q.S. Liu, Y.F. Jiang, K. Jin, et al., Sci. Bull. 65 (2020) 272–275. doi: 10.1016/j.scib.2020.01.001

  9. [9]

     C.K. Sun, F. Pan, H.J. Bin, et al., Nat. Commun. 9 (2018) 743. doi: 10.1038/s41467-018-03207-x

  10. [10]

      A.L. Tang, P.Q. Cong, T.T. Dai, Z.T. Wang, E.J. Zhou, Adv. Mater. 35 (2023) 2300175.

  11. [11]

      S.S. Li, L. Ye, W.C. Zhao, et al., J. Am. Chem. Soc. 140 (2018) 7159–7167. doi: 10.1021/jacs.8b02695

  12. [12]

      K. Chong, X.P. Xu, H.F. Meng, et al., Adv. Mater. 34 (2022) 2109516. doi: 10.1002/adma.202109516

  13. [13]

      L. Zhu, M. Zhang, J.Q. Xu, et al., Nat. Mater. 21 (2022) 656–663. doi: 10.1038/s41563-022-01244-y

  14. [14]

      R. Zeng, L. Zhu, M. Zhang, et al., Nat. Commun. 14 (2023) 4148. doi: 10.1038/s41467-023-39832-4

  15. [15]

      X.J. Li, F.S. Pan, C.K. Sun, et al., Nat. Commun. 10 (2019) 519. doi: 10.1038/s41467-019-08508-3

  16. [16]

      W.X. Wang, Y. Cui, T.B. Zhang, et al., Joule 7 (2023) 1067–1079. doi: 10.1016/j.joule.2023.04.003

  17. [17]

      L. Xie, W. Song, J.F. Ge, et al., Nano Energy 82 (2021) 105770. doi: 10.1016/j.nanoen.2021.105770

  18. [18]

      X.B. Zhou, H.B. Wu, U. Bothra, et al., Mater. Horiz. 10 (2023) 566–575. doi: 10.1039/d2mh01229d

  19. [19]

      F.J. Bai, J.Q. Zhang, A.P. Zeng, et al., Joule 5 (2021) 1231–1245. doi: 10.1016/j.joule.2021.03.020

  20. [20]

      Z.T. Wang, A.L. Tang, H.L. Wang, et al., Chem. Eng. J. 451 (2023) 139080. doi: 10.1016/j.cej.2022.139080

  21. [21]

      B. Xiao, A.L. Tang, J. Yang, Z.X. Wei, E.J. Zhou, ACS Macro Lett. 6 (2017) 410–414. doi: 10.1021/acsmacrolett.7b00097

  22. [22]

      B. Xiao, A.L. Tang, J.Q. Zhang, et al., Adv. Energy Mater. 7 (2017) 1602269. doi: 10.1002/aenm.201602269

  23. [23]

      A.L. Tang, B. Xiao, Y.M. Wang, et al., Adv. Funct. Mater. 28 (2018) 1704507. doi: 10.1002/adfm.201704507

  24. [24]

      A. Tang, W. Song, B. Xiao, et al., Chem. Mater. 31 (2019) 3941–3947. doi: 10.1021/acs.chemmater.8b05316

  25. [25]

      T.T. Dai, A.L. Tang, Z.H. He, et al., Energy Environ. Sci. 16 (2023) 2199–2211. doi: 10.1039/d3ee00344b

  26. [26]

      Y. Cui, Y. Xu, H.F. Yao, et al., Adv. Mater. 33 (2021) 2102420. doi: 10.1002/adma.202102420

  27. [27]

      Y. Yang, Z.H. Chen, T.Q. Wang, et al., Org. Electron. 98 (2021) 106289. doi: 10.1016/j.orgel.2021.106289

  28. [28]

      Y. Xu, Y. Cui, H.F. Yao, et al., Adv. Mater. 33 (2021) 2101090. doi: 10.1002/adma.202101090

  29. [29]

      Q. Wu, Y. Yu, X.X. Xia, et al., Joule 6 (2022) 2138–2151. doi: 10.1016/j.joule.2022.07.001

  30. [30]

      Q.P. Fan, W.Y. Su, Y. Wang, et al., Sci. China Chem. 61 (2018) 531–537. doi: 10.1007/s11426-017-9199-1

  31. [31]

      J.M. Cao, L.F. Yi, L.M. Ding, J. Semicond. 43 (2022) 030202. doi: 10.1088/1674-4926/43/3/030202

  32. [32]

      Y.X. Li, J.D. Lin, X.Z. Che, et al., J. Am. Chem. Soc. 139 (2017) 17114–17119. doi: 10.1021/jacs.7b11278

  33. [33]

      X. Liu, Y.X. Li, K. Ding, S. Forrest, Phys. Rev. Appl. 11 (2019) 024060. doi: 10.1103/PhysRevApplied.11.024060

  34. [34]

      H. Zhang, H.F. Yao, J.X. Hou, et al., Adv. Mater. 30 (2018) 1800613. doi: 10.1002/adma.201800613

  35. [35]

      J.F. Huang, S.S. Li, J.Z. Qin, et al., ACS Appl. Mater. Interfaces 13 (2021) 45806–45814. doi: 10.1021/acsami.1c11412

  36. [36]

      Y.X. Tang, H.X. Feng, Y.Y. Liang, et al., ACS Appl. Polym. Mater. 5 (2021) 2298–2306.

  37. [37]

      X.C. Wang, A.L. Tang, J. Yang, et al., Sci. China Chem. 63 (2020) 1666–1674. doi: 10.1007/s11426-020-9840-x

  38. [38]

      K.Y. Zuo, T.T. Dai, Q. Guo, et al., ACS Appl. Energy Mater. 5 (2022) 14271–14279. doi: 10.1021/acsaem.2c02823

  39. [39]

      H.F. Yao, D.P. Qian, H. Zhang, et al., Chin. J. Chem. 36 (2018) 491–494. doi: 10.1002/cjoc.201800015

  40. [40]

      X.G. Li, Z.T. Wang, A.L. Tang, et al., Macromol. Rapid Commun. 44 (2023) 2300019. doi: 10.1002/marc.202300019

  41. [41]

      C. Zhu, J. Yuan, F.F. Cai, et al., Energy Environ. Sci. 13 (2020) 2459–2466. doi: 10.1039/d0ee00862a

  42. [42]

      S.R. Cowan, A. Roy, A.J. Heeger, Phys. Rev. B 82 (2010) 245207. doi: 10.1103/PhysRevB.82.245207

  43. [43]

      Y.F. Wang, J. Li, T.F. Li, et al., Small 15 (2019) 1903977. doi: 10.1002/smll.201903977

  44. [44]

      Z. Chen, X. Chen, Z.Y. Jia, et al., Joule 5 (2021) 1832–1844. doi: 10.1016/j.joule.2021.04.002

  45. [45]

      F. Urbach, Phys. Rev. 92 (1953) 1324-1324.

  46. [46]

      S. Liu, J. Yuan, W.Y. Deng, et al., Nat. Photonics 14 (2020) 300–305. doi: 10.1038/s41566-019-0573-5

  47. [47]

      Y.G. Cui, P.P. Zhu, H.W. Hu, et al., Angew. Chem. Int. Ed. 62 (2023) e202304931. doi: 10.1002/anie.202304931

  48. [48]

      P.Y. Zhang, Z.Y. Zhang, H. Sun, et al., Chin. Chem. Lett. 35 (2024) 108802. doi: 10.1016/j.cclet.2023.108802

  49. [49]

      J.K. Zhang, B. Yu, Y.P. Sun, H.Z. Yu, Adv. Energy Mater. 13 (2023) 2300382. doi: 10.1002/aenm.202300382

  50. [50]

      Z.P. Yu, X. Li, C.L. He, et al., Chin. Chem. Lett. 31 (2020) 1991–1996. doi: 10.1016/j.cclet.2019.12.003

  51. [51]

      N. An, Y.H. Cai, H.B. Wu, et al., Adv. Mater. 32 (2020) 2002122. doi: 10.1002/adma.202002122

  52. [52]

      H.J. Li, S.Q. Liu, X.T. Wu, et al., Energy Environ. Sci. 15 (2022) 2130–2138. doi: 10.1039/d2ee00639a

• 

  Figure 1  Chemical structures of BTA5, BTA5-Cl and BTA5-2Cl (a) and the polymer D18-Cl (b). The absorption spectra of donor and three acceptors in chloroform solution (c) and neat films (d). (e) Energy levels diagram of D18-Cl, BTA5, BTA5-Cl and BTA5-2Cl.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  J-V curves (a) and EQE spectra (b) of the OSCs based on D18-Cl:BTA5, D18-Cl:BTA5-Cl and D18-Cl:BTA5-2Cl. (c) The PL of neat films and blend films with excitation at 650 nm. J^1/2-V curves of electron-only (d) and hole-only (e) mobilities tested by the SCLC method.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Atomic force microscopy height images of D18-Cl:BTA5 (a), D18-Cl:BTA5-Cl (b) and D18-Cl:BTA5-2Cl (c). 2D GIWAXS patterns of D18-Cl:BTA5 (d), D18-Cl:BTA5-Cl (e), D18-Cl:BTA5–2Cl (f) and the corresponding 1D profiles (g).

  下载: 全尺寸图片 幻灯片

  Table 1.  Photovoltaic parameters of the optimized three OSCs.

  下载: 导出CSV
•