Banana‐shaped electron acceptors with an electron‐rich core fragment and 3D packing capability

Abstract

The emergence of Y6‐type nonfullerene acceptors has greatly enhanced the power conversion efficiency (PCE) of organic solar cells (OSCs). However, which structural feature is responsible for the excellent photovoltaic performance is still under debate. In this study, two Y6‐like acceptors BDOTP‐1 and BDOTP‐2 were designed. Different from previous Y6‐type acceptors featuring an A–D–Aʹ–D–A structure, BDOTP‐1, and BDOTP‐2 have no electron‐deficient Aʹ fragment in the core unit. Instead, there is an electron‐rich dibenzodioxine fragment in the core. Although this modification leads to a marked change in the molecular dipole moment, electrostatic potential, frontier orbitals, and energy levels, BDOTP acceptors retain similar three‐dimensional packing capability as Y6‐type acceptors due to the similar banana‐shaped molecular configuration. BDOTP acceptors show good performance in OSCs. High PCEs of up to 18.51% (certified 17.9%) are achieved. This study suggests that the banana‐shaped configuration instead of the A–D–Aʹ–D–A structure is likely to be the determining factor in realizing high photovoltaic performance.

INTRODUCTION

In 2019, Yuan et al.¹ first reported a nonfullerene acceptor Y6. Then, the power conversion efficiency (PCE) of organic solar cells (OSCs) increased rapidly from <15% to >18%.^2,3 Among Y6-based OSCs, D18:Y6 cells delivered the highest efficiency of 18.22%.⁴ Focusing on the Y6 structure, scientists have performed a lot of modifications, such as changing the backbone units and side chains, incorporating heteroatoms (e.g., Se, N, O), and introducing asymmetric structures.^5–26 These efforts have produced many highly efficient Y6 analogs. Among them, BTP-eC9, N3, and L8-BO have shown improvement in the PCE to almost 19%.^13,27–29

Different from traditional acceptor–donor–acceptor (A–D–A) nonfullerene acceptors, Y6-type acceptors have two unique structural features. The first feature is the electron-deficient fragment (e.g., benzothiadiazole, benzoselenadiazole, benzotriazole, quinoxaline) in the core unit. In this regard, Y6-type acceptors are also called A–D–Aʹ–D–A acceptors. There is a view that the electron-deficient Aʹ fragment fine-tunes the molecular orbitals and energy levels, thus improving the light absorption and electron affinity of the acceptors and leading to high photovoltaic performance.^1,30 The second feature is the banana-shaped backbone. Unlike the linear backbone in traditional A–D–A acceptors, the banana-shaped backbone allows Y6-type acceptors to pack via multiple modes, such as end-to-end packing, end-to-core packing, and core-to-core packing.³¹ Some studies pointed out that the multiple packing enhances the electron coupling between the molecules and facilitates the formation of a continuous three-dimensional (3D) network in the crystal.³² It confers Y6-type acceptors with strong charge-transporting capability. To date, it remains an open question whether the high performance of Y6-type acceptors is determined by the A–D–Aʹ–D–A structure or the banana-shaped configuration, or both.

In this study, we designed and synthesized two new Y6-like nonfullerene acceptors: BDOTP-1 and BDOTP-2 (Figure 1). Different from previous A–D–Aʹ–D–A Y6-type acceptors, BDOTP-1 and BDOTP-2 have no electron-deficient fragment in the core unit. Instead, there is an electron-rich dibenzodioxine fragment in the core. Meanwhile, BDOTP-1 and BDOTP-2 retain the banana-shaped molecular configuration like Y6. Although there are marked differences in the molecular dipole moment, electrostatic potential (ESP), frontier orbitals, and energy levels between Y6 and the new acceptors, BDOTP-1 and BDOTP-2 show similar packing behavior and crystal structure as Y6-type acceptors and high performance in solar cells. On blending with a polymer donor D18-B,²⁸ BDOTP-1 and BDOTP-2 delivered PCEs of 16.93% and 15.48%, respectively. Moreover, a PCE of 18.51% (certified 17.9%) was achieved in D18-B:Y6:PBDOTP-1 ternary cells.

Figure 1. Chemical structures of Y6, BDOTP-1, and BDOTP-2.

RESULTS AND DISCUSSION

The synthetic route for BDOTP acceptors is shown in Scheme 1. Stille coupling of (6-undecylthieno[3,2-b]thiophen-2-yl)tributylstannane or (6-(2-butyloctyl)thieno[3,2-b]thiophen-2-yl)trimethylstannane with 1,4-dibromo-2,3-difluoro-5,6-dinitrobenzene yielded compound 1a or 1b in 68% or 53% yield, respectively. Treatment of compound 1a or 1b with catechol under basic conditions afforded the annulation product 2a or 2b in 82% or 47% yield, respectively. Then, compound 2a or 2b underwent consecutive Cadogan cyclization,³³ alkylation, and the Vilsmeier reaction, and produced compound 3a or 3b in 21% or 19% yield, respectively. Finally, Knoevenagel condensation of compound 3a or 3b with 2-(5,6-difluoro-3-oxo-2,3-dihydro-1H-inden-1-ylidene)malononitrile afforded BDOTP-1 or BDOTP-2 in 81% or 94% yield, respectively. All new compounds were characterized by nuclear magnetic resonance and mass spectra. Single crystals of BDOTP-2 were obtained by slow diffusion of CH₃CN into the CHCl₃ solution of BDOTP-2. The molecular structure of BDOTP-2 was confirmed by single-crystal X-ray diffraction (XRD) analysis (vide infra).

Scheme 1. The synthetic route for BDOTP-1 and BDOTP-2.

We performed density functional theory (DFT) calculations to obtain information about the molecular configuration and electronic structures (Figure 2). Molecular models of Y6 and BDOTP acceptors were constructed and optimized at the B3LYP/6-31G(d) level. To simplify the calculation, the outside alkyl chains were replaced by methyl groups, and the inside alkyl chains were replaced by isobutyl groups. The optimized geometries indicate similar banana-shaped backbones of Y6 and BDOTP acceptors. However, the molecular dipole moment and ESP distribution are quite different. Compared with Y6 with a dipole moment of 1.51 D, BDOTP acceptors have a larger dipole moment of 2.59 D, pointing to the opposite direction. The ESP distributions of Y6 and BDOTP acceptors were mapped on their van der Waals surfaces (Figure 2B).^34,35 For Y6, the negative ESP not only distributes on the rim of the end units (i.e., cyano, carbonyl and F groups) but also on the rim of the core (i.e., ESP on thiodiazole moiety). However, for BDOTP acceptors, the negative ESP mainly distributes on the rim of the end units. The ESP distribution reflects the electron-deficient nature of the BT fragment in Y6 and the electron-rich nature of the BDBO fragment in BDOTP acceptors. The lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO) for the molecular models are shown in Figure 2C. Interestingly, the LUMO distributions are similar for Y6 and BDOTP acceptors, while the HOMO distributions are different. For Y6, HOMO distributes on either the core or the end units. For BDOTP acceptors, HOMO only distributes on the core unit (mainly on the BDBO fragment). Compared with Y6, both LUMO and HOMO energy levels for BDOTP acceptors increased. Based on the above theoretical studies, it can be concluded that BDOTP acceptors are quite different from Y6 in the molecular dipole moment, ESP, frontier orbitals, and energy levels.

Figure 2. (A) The optimized configuration and dipole moment, (B) ESP, and (C) frontier orbitals and energy levels for Y6 (left) and BDOTP acceptors (right). ESP, electrostatic potential; HOMO, highest occupied molecular orbital; LUMO, lowest unoccupied molecular orbital.

The absorption spectra for BDOTP-1, BDOTP-2, and Y6 are shown in Figure 3. In solution, the spectra for BDOTP-1 and BDOTP-2 are almost identical due to their identical conjugation systems. A peak at 737 nm was observed for BDOTP-1 and BDOTP-2. Compared with Y6, with a peak at 732 nm, BDOTP acceptors show red-shifted spectra in solution. This could be due to the electron-donating core of BDOTP acceptors, which enhances the push–pull effect between the core and the end units, thus leading to a red shift in the spectra. The absorption spectra for BDOTP-1 and BDOTP-2 films show a red shift as compared to their solution ones, suggesting enhanced J-aggregation in films. Compared with BDOTP-2, BDOTP-1 shows a larger red shift in the film. The absorption peaks for BDOTP-1 and BDOTP-2 films occur at 802 and 783 nm, respectively. The spectra for BDOTP-1 and BDOTP-2 films are less red-shifted as compared to that of the Y6 film. This suggests weaker J-aggregation tendency of BDOTP-1 and BDOTP-2 than Y6.³¹ The estimated optical band gaps (E_g^opt) from the absorption onset of the films are 1.42, 1.45, and 1.36 eV for BDOTP-1, BDOTP-2, and Y6, respectively. The electrochemical properties of BDOTP acceptors were studied using the cyclic voltammetry method,³⁶ with Y6 as the reference (Figure S17). The LUMO levels estimated from the onset potentials of reduction (E_on^red) are −3.85, −3.81, and −3.88 eV for BDOTP-1, BDOTP-2, and Y6, respectively. Both BDOTP-1 and BDOTP-2 show higher LUMO levels than Y6, consisting with the DFT calculation.

Figure 3. The absorption spectra for BDOTP-1, BDOTP-2, and Y6 (A) in CHCl3 and (B) as films.

The molecular configurations and packing modes for BDOTP-2 derived from single-crystal XRD analysis are shown in Figure 4. In Figure 4A, the top view indicates the banana-shaped configuration of BDOTP-2. The side view shows good planarity of the backbone. Owing to the steric hindrance, the two inside 2-ethylhexyl chains are located above and below the backbone. In the crystal, two adjacent molecules form two types of dimers (Figure 4B). One dimer features end-to-core packing and the other one shows end-to-end packing. The two packing modes are frequently observed in the crystals of Y6-type acceptors,^{13,17,21,32,37–39} suggesting the common feature of banana-shaped backbones. BDOTP-2 molecules may create a 3D network. On viewing along the c axis, the stacked backbones of BDOTP-2 present a honeycomb-like structure (Figure 4C). The honeycomb holes are filled with alkyl chains. The holes have a length of 18.6 Å and a width of 13.8 Å. On viewing along the b axis, π–π stacking d-spacings of 3.17 (end-to-core packing) and 3.37 Å (end-to-end packing) were determined, suggesting a strong intermolecular interaction (Figure 4D). The π-plane overlapping in end-to-core packing can strengthen the interaction, thus leading to shorter π–π stacking d-spacing. The packing in the film was further investigated by grazing-incidence wide-angle X-ray scattering (GIWAXS). The pure Y6 film shows a (010) peak in the out-of-plane (OOP) direction, with characteristic (110) and (11$$\overline{){\rm{l}}}$$) peaks in the in-plane (IP) direction (Figure S18A), consistent with a previous report.¹³ Because of the relatively low solubility of BDOTP-1, many crystalline aggregates (diameter ∼20 μm) exist in the film (Figure S18B). Therefore, the GIWAXS pattern of BDOTP-1 presents a polycrystalline diffraction similar to that of Y6, suggesting that BDOTP-1 and Y6 have similar packing. The GIWAXS pattern of BDOTP-2 (Figure S18C) is very similar to that of L8-BO.¹³ We observe a (010) peak in the OOP direction, a (021) peak in the 65°-tilted OOP direction, and an (11$$\bar{{\rm{l}}}$$) peak in the IP direction. It should be noted that BDOTP-1 and Y6 have the same set of alkyl chains, while BDOTP-2 and L8-BO have the same set of alkyl chains (Figure S19). Despite different core units, because of the similar banana-shaped backbone and the same set of alkyl chains, BDOTP acceptors have similar packing to Y6-type acceptors. The electron-transporting property of BDOTP acceptors was studied using the space-charge limited current (SCLC) method (Figure S20, Table S1).^40–44 BDOTP-1 and BDOTP-2 present high electron mobility (μ_e) values of 9.10 × 10⁻⁴ and 5.60 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively, indicating that they could be good acceptors in bulk-heterojunction solar cells. The 3D packing of BDOTP may account for the good electron-transporting property.

Figure 4. (A) Top view (left) and side view (right) for the single-crystal structure of BDOTP-2; (B) two packing modes in the crystal; (C) the molecular packing viewed along the c axis (alkyl chains are omitted for clarity); and (D) the molecular packing viewed along the b axis (alkyl chains are omitted for clarity).

The photovoltaic performance of BDOTP-1 and BDOTP-2 was evaluated by creating OSCs with the structure ITO/PEDOT:PSS/D18-B:acceptor/PDIN/Ag. The D/A ratio, active layer thickness, and additive (1-chloronaphthalene) content were optimized (Tables S2–S7). J–V curves and external quantum efficiency (EQE) spectra for the best solar cells are shown in Figure 5, and the performance parameters are listed in Table 1. With a D/A ratio of 1:1 and an active layer thickness of 114 nm, the D18-B:BDOTP-1 cells yielded the highest PCE of 16.93%, with an open-circuit voltage (V_oc) of 0.938 V, a short-circuit current density (J_sc) of 24.92 mA cm⁻², and a fill factor (FF) of 72.4%. With a D/A ratio of 1:1 and an active layer thickness of 101 nm, the D18-B:BDOTP-2 cells yielded the highest PCE of 15.48%, with a V_oc of 0.977 V, a J_sc of 22.29 mA cm⁻², and an FF of 71.1%. The addition of 1-chloronaphthalene to the blend led to deterioration of device performance for both cells. BDOTP-1 showed superior performance than BDOTP-2 due to higher J_sc and FF. The higher J_sc of BDOTP-1 cells resulted from the higher and broader EQE response (Figure 5B). The highest EQE (EQE_max) values for BDOTP-1 and BDOTP-2 cells were 82% and 80%, respectively, suggesting more efficient charge generation in BDOTP-1 cells. We investigated the exciton dissociation probability (P_diss) and bimolecular recombination in BDOTP-1 and BDOTP-2 cells (Figure S21). The BDOTP-1 cells showed a higher P_diss and an α value closer to 1, suggesting more efficient exciton dissociation and less bimolecular recombination in the active layer. We also studied charge carrier mobility of the active layer using the SCLC method (Figures S22 and S23). The D18-B:BDOTP-1 film showed higher μ_e and higher hole mobility (μ_h) than the D18-B:BDOTP-2 film (Table S9). The better charge transport and suppressed charge recombination in the D18-B:BDOTP-1 film account for its higher J_sc and FF. The morphology and molecular packing for D18-B:BDOTP-1 and D18-B:BDOTP-2 films were studied by atomic force microscopy (AFM) and GIWAXS, respectively (Figure 5C,D). AFM images showed clear nanofiber structures (diameter ∼20 nm) in both films. The D18-B:BDOTP-1 film was smoother than the D18-B:BDOTP-2 film, and their root-mean-square roughness (R_rms) values were 1.16 and 1.46 nm, respectively. For both the D18-B:BDOTP-1 and D18-B:BDOTP-2 films, the GIWAXS pattern presented a strong (010) π–π stacking peak in the OOP direction and a strong (100) lamellar stacking peak in the IP direction, suggesting that donor and acceptor molecules prefer a “face-on” orientation in the blend films. For the D18-B:BDOTP-2 film, a peak in the 65°-tilted OOP direction was also observed. This is the (021) peak of BDOTP-2. The π–π stacking and lamellar stacking d-spacings were 3.69 and 21.25 Å for the D18-B:BDOTP-1 film, and 3.71 and 21.09 Å for the D18-B:BDOTP-2 film, respectively (Table S10). Using the Scherrer equation, the corresponding crystallite coherence lengths (CCL) for the (010) and (100) peaks were calculated to be 15.92 and 76.05 Å for the D18-B:BDOTP-1 film, and 15.02 and 76.92 Å for the D18-B:BDOTP-2 film, respectively.⁴⁵ The shorter π–π stacking d-spacing and the larger CCL₀₁₀ for the D18-B:BDOTP-1 film facilitate charge transport, and are responsible for the higher μ_e and μ_h in the D18-B:BDOTP-1 film.

Figure 5. (A) J–V curves for D18-B:BDOTP-1 (1:1) and D18-B:BDOTP-2 (1:1) solar cells; (B) EQE spectra for D18-B:BDOTP-1 (1:1) and D18-B:BDOTP-2 (1:1) solar cells; (C) AFM height and phase images and the GIWAXS pattern for D18-B:BDOTP-1 (1:1) blend film; (D) AFM height and phase images and the GIWAXS pattern for D18-B:BDOTP-2 (1:1) blend film. AFM, atomic force microscope; EQE, external quantum efficiency; GIWAXS, grazing-incidence wide-angle X-ray scattering.

Table 1 Performance data of the solar cells

Active layer	Voc (V)	Jsc (mA/cm2)	FF (%)	PCE (%)
D18-B:BDOTP-1 (1:1)	0.938	24.92	72.4	16.93 (16.76)a
D18-B:BDOTP-2 (1:1)	0.977	22.29	71.1	15.48 (15.31)
D18-B:Y6 (1:1.4)	0.853	26.45	78.6	17.74 (17.69)
D18-B:Y6:BDOTP-1 (1:1.2:0.2)	0.859	27.74	77.6	18.51 (18.45)

Abbreviations: FF, fill factor; PCE, power conversion efficiency.

a

The data in the parentheses represent averages for 10 cells.

Though PCEs for BDOTP acceptors were lower than that of Y6 (Table 1), they yielded higher V_oc (0.09–0.12 V increase). This implies that they may reduce the energy loss (E_loss).^46,47 We tried to use BDOTP-1 as the third component to improve the V_oc of D18-B:Y6 cells and to achieve higher PCEs. When the D18-B:Y6:BDOTP-1 ratio was 1:1.2:0.2, the ternary cells yielded a higher PCE of 18.51% than D18-B:Y6 binary cells (PCE of 17.74%), with a V_oc of 0.859 V, a J_sc of 27.74 mA cm⁻², and an FF of 77.6% (Table S8). The best ternary cells were also measured at the National Institute of Metrology, and a certified PCE of 17.9% (V_oc, 0.853 V; J_sc, 26.74 mA cm⁻²; FF, 78.3%; effective area, 2.580 mm²) was recorded (Figure S24). Addition of a small amount of BDOTP-1 to the D18-B:Y6 blend effectively enhanced V_oc and J_sc. The increase in J_sc originates from the higher EQE response in 380–850 nm (Figure 6). The EQE_max values for D18-B:Y6 cells and D18-B:Y6:BDOTP-1 cells were 85% and 89%, respectively. The P_diss and α values were 98.1% and 0.983 for the binary cells, and 98.5% and 0.985 for the ternary cells, respectively, suggesting that BDOTP-1 promotes exciton dissociation and reduces bimolecular recombination (Figure S25). The μ_h and μ_e values obtained from SCLC measurements were 5.61 × 10⁻⁴ and 6.76 × 10⁻⁴ cm² V⁻¹ s⁻¹ for the binary cells, and 8.74 × 10⁻⁴ and 9.23 × 10⁻⁴ cm² V⁻¹ s⁻¹ for the ternary cells, respectively (Figures S26 and S27, Table S11), indicating that BDOTP-1 improves charge transport in the active layer. The more efficient exciton dissociation and charge transport, and reduced charge recombination led to higher J_sc in D18-B:Y6:BDOTP-1 cells. The morphology and molecular packing for binary and ternary blend films were investigated. The two films showed similar AFM images and GIWAXS patterns (Figure 6C,D). In AFM images, D18-B:Y6 and D18-B: Y6:BDOTP-1 films presented finer nanofibers (diameter ∼16 nm) than D18-B:BDOTP-1 and D18-B:BDOTP-2 films. These morphologies could explain the higher J_sc value for D18-B:Y6 and D18-B:Y6:BDOTP-1 cells as compared to D18-B:BDOTP-1 and D18-B:BDOTP-2 cells, since they provided more donor/acceptor interfaces for exciton dissociation. In GIWAXS patterns, D18-B:Y6 and D18-B:Y6:BDOTP-1 films showed similar d-spacings and CCLs for both π–π stacking and lamella stacking (Table S12). A small amount of BDOTP-1 in the D18-B:Y6 film slightly affected the morphology and molecular packing.

Figure 6. (A) J–V curves for D18-B:Y6 (1:1.4) and D18-B:Y6:BDOTP-1 (1:1.2:0.2) solar cells; (B) EQE spectra for D18-B:Y6 (1:1.4) and D18-B:Y6:BDOTP-1 (1:1.2:0.2) solar cells; (C) AFM height and phase images and the GIWAXS pattern for the D18-B:Y6 (1:1.4) blend film; and (D) AFM height and phase images and the GIWAXS pattern for the D18-B:Y6:BDOTP-1 (1:1.2:0.2) blend film. AFM, atomic force microscopy; EQE, external quantum efficiency; GIWAXS, grazing-incidence wide-angle X-ray scattering.

To understand why D18-B:BDOTP-1 and D18-B:BDOTP-2 cells showed high V_oc and why D18-B:Y6:BDOTP-1 ternary cells yielded higher V_oc than D18-B:Y6 binary cells, we performed E_loss analysis for these cells according to a method in the literature.⁴⁸ The band gaps (E_gap) for four cells were extracted from the EQE spectra (Figure S28).⁴⁹ The E_gap values were 1.465, 1.492, 1.402, and 1.402 eV for D18-B:BDOTP-1, D18-B:BDOTP-2, D18-B:Y6, and D18-B:Y6:BDOTP-1, respectively. The total energy loss was calculated according to the equation E_loss = E_gap − qV_oc. As summarized in Table 2, BDOTP acceptors indeed led to smaller E_loss in binary and ternary solar cells. D18-B:BDOTP-2 cells yielded the smallest E_loss of 0.515 eV, which is 0.034 eV lower than that of D18-B:Y6 cells. Next, we quantitatively analyzed the E_loss components in depth.⁵⁰ E_loss includes three parts: [Image Omitted. See PDF]

Table 2 E_loss analysis for the solar cells

Active layer	Egap (eV)	qVoc (eV)	Eloss (eV)	ΔE1 (eV)	ΔE2 (eV)	ΔE3 (eV)
D18-B:BDOPT-1	1.465	0.938	0.527	0.267	0.067	0.196
D18-B:BDOPT-2	1.492	0.977	0.515	0.268	0.062	0.196
D18-B:Y6	1.402	0.853	0.549	0.262	0.067	0.222
D18-B:Y6:BDOPT-1	1.402	0.868	0.534	0.262	0.066	0.209

ΔE₁ is the radiative recombination loss above the band gap $$({\rm{\Delta }}{E}_{1}={E}_{\text{gap}}-{\rm{q}}{V}_{\mathrm{oc}}^{\mathrm{SQ}},{V}_{\mathrm{oc}}^{\mathrm{SQ}}$$ is the Shockley–Queisser limit output voltage). For all solar cells, ΔE₁ is unavoidable and is only related to E_gap (Equation S2). In our case, D18-B:BDOTP-1 and D18-B:BDOTP-2 cells showed slightly larger ΔE₁ (0.267 and 0.268 eV) than D18-B:Y6 and D18-B:Y6:BDOTP-1 cells (0.262 eV) due to their larger E_gap. ΔE₂ is the radiative recombination loss below the band gap, which can be obtained from Fourier-transform photocurrent spectroscopy–EQE measurements (Figure S29, Equation S3). It was found that the difference in ΔE₂ was small for the four cells (Table 2). ΔE₃ is the nonradiative recombination loss, which can be calculated from the EQE of electroluminescence (EQE_EL) for the solar cells (Figure 7, Equation S4). The higher the EQE_EL, the smaller the ΔE₃. As shown in Figure 7, the EQE_EL (1 V) value for both BDOTP-1 and BDOTP-2 binary cells is 5 × 10⁻⁴, which is higher than that of Y6 binary cells (2 × 10⁻⁴). The ΔE₃ values for D18-B:BDOTP-1 and D18-B:BDOTP-2 cells (0.196 eV) were much smaller than that of D18-B:Y6 cells (0.222 eV). Addition of a small amount of BDOTP-1 to the D18-B:Y6 blend increased the EQE_EL to 3 × 10⁻⁴, thus leading to a smaller ΔE₃ of 0.209 eV for the ternary cells. From the above analysis, it can be concluded that BDOTP acceptors have stronger capability in reducing nonradiative recombination loss than Y6. The high V_oc values of D18-B:BDOTP-1 and D18-B:BDOTP-2 cells stem from the larger E_gap and the lower ΔE₃. The higher V_oc of D18-B:Y6:BDOTP-1 cells than that of D18-B:Y6 cells is due to the reduced ΔE₃.

Figure 7. EQEEL–voltage plots for D18-B:BDOTP-1, D18-B:BDOTP-2, D18-B:Y6, and D18-B:Y6:BDOTP-1 solar cells. EQEEL, external quantum efficiency of electroluminescence.

CONCLUSION

We developed two banana-shaped nonfullerene acceptors BDOTP-1 and BDOTP-2. Different from previous A–D–Aʹ–D–A Y6-type acceptors, BDOTP acceptors have an electron-rich fragment in the core. This modification leads to different molecular dipole moments, ESPs, frontier orbitals, and energy levels. BDOTP acceptors retain similar 3D packing as Y6-type acceptors due to the similar banana-shaped configuration. Due to the good electron-transporting property and the strong capability of suppressing nonradiative recombination, BDOTP-1 and BDOTP-2 are efficient acceptors. PCEs of 16.93% and 18.51% have been achieved in BDOTP-1-based binary and ternary solar cells, respectively. This study suggests that the high performance of Y6-type acceptors might be predominantly dependent on the banana-shaped configuration instead of the A–D–Aʹ–D–A structure. This finding will shed light on designing highly efficient nonfullerene acceptors.

ACKNOWLEDGMENTS

The authors are grateful for the open research fund of the Songshan Lake Materials Laboratory (2021SLABFK02), the National Key Research and Development Program of China (2017YFA0206600), and the National Natural Science Foundation of China (51922032 and 21961160720).

CONFLICTS OF INTEREST

The authors declare no conflicts of interest.