The impact of axisymmetric and centrosymmetric molecular architectures in non-fused ring electron acceptors on photovoltaic performance

Abstract

The geometric configuration of electron acceptors significantly governs molecular dipole moments and stacking behavior, thereby critically influencing power conversion efficiencies (PCEs) in organic solar cells (OSCs). In this study, we designed and synthesized two non-fused ring electron acceptors (NFREAs), TTCIC (axisymmetric) and TCIC (centrosymmetric), by incorporating 3,6-bis(octan-3-yloxy)thieno[3,2-b]thiophene and 3,4-bis(octan-3-yloxy)thiophene units, respectively. Compared to TTCIC, TCIC exhibits a higher LUMO (−3.89 eV vs. −3.98 eV), a lower HOMO (−5.40 eV vs. −5.35 eV), a large dipole moment change (0.217 D vs. 0 D) between the ground state and excited state dipoles, and weaker intermolecular interactions. Interestingly, both acceptors showed an edge-on molecular orientation in the films; however, after blending with a polymer donor, PBDB-T, TTCIC blend films exhibited preferential edge-on molecular alignment, whereas TCIC blend films adopted a face-on orientation. This morphological contrast induced stronger charge carrier recombination in PBDB-T:TTCIC blends. Consequently, PBDB-T:TTCIC-based OSCs achieved an exceptionally low PCE of 0.60%, while PBDB-T:TCIC devices delivered a moderate PCE of 8.66%. These results demonstrate that fine-tuning of NFREA geometric configurations is essential for optimizing the molecular stacking orientation and enhancing the OSC performance.

---

Cunbin An

Cunbin An received his PhD degree from the Max Planck Institute for Polymer Research (MPIP) in 2015. After carrying out postdoctoral research in the same group, he joined the Institute of Chemistry, Chinese Academy of Sciences (ICCAS) in 2016. In 2022, he moved to Capital Normal University (CNU), where he is now a Full Professor. His current research focuses on the development of novel organic functional materials, as well as the investigation of new mechanisms for application in organic electronic devices.

Introduction

With inherent advantages including mechanical flexibility, lightweight, transparency, and cost-effective fabrication processes, bulk heterojunction (BHJ) organic solar cells (OSCs) have emerged as competitive candidates among various photovoltaic technologies.^1–6 The fundamental architecture of OSCs comprises electron donor and acceptor materials. Among these components, fused ring electron acceptors (FREAs) have become a predominant research focus due to their potential to deliver superior power conversion efficiencies (PCEs).^7–10 A pivotal breakthrough occurred with the development of ITIC, a centrosymmetric acceptor, that first demonstrated the capability of FREAs to surpass traditional fullerene-based acceptors in terms of PCEs of OSCs.^7,11 In recent years, OSCs incorporating axisymmetric Y6 and its derivatives have achieved outstanding PCEs exceeding 20%.^12–23 These FREAs typically exhibit distinctive characteristics including intense near-infrared absorption, substantial dipole moments, and optimal crystallinity. Therefore, the geometric configuration of FREAs play a significant role in the rapid advancement of OSCs.

Taking advantage of the chemical tailoring of FREAs, it is easy to modify the molecular geometric configuration by extending the conjugation length, which influences electron distribution across the conjugated backbone, thereby affecting absorption spectra, molecular energy levels, and molecular stacking behaviors.^24,25 For instance, Ge and coworkers developed three FREAs with distinct geometries, TB-4Cl, PTIC-4Cl, and PT2IC-4Cl, by systematically extending the conjugation length.²⁶ As the conjugated length increases, the optical bandgap of these materials gradually decreases. Meanwhile, the energy level of the lowest unoccupied molecular orbital (LUMO) changes slightly, while that of the highest occupied molecular orbital (HOMO) moves significantly upwards. Among them, PTIC-4Cl exhibited a larger dipole moment and the strongest π–π stacking distance compared to TB-4Cl and PT2IC-4Cl, leading to enhanced PCEs from 12.51% to 14.81%. Similarly, Zhang and coworkers reported two FREAs with different configurations, IDTP-4F (S-shape) and IDTTP-4F (C-shape), via fusing an additional thiophene unit into the conjugated backbone.²⁷ Interestingly, both FREAs exhibited remarkably similar molecular energy levels and absorption spectra. However, the C-shaped acceptor demonstrated stronger aggregation, forming larger domains in the blend film, resulting in a poorer PCE of 12.6%. In contrast, the S-shaped acceptor showed a favorable BHJ morphology, yielding a better PCE of 14.3%. Although extending the conjugation length by fusing an aromatic ring could enhance PCEs in OSCs, this process usually produces excessively large fused-ring structures, which complicate synthesis and reduce solubility. This frequently leads to severe aggregation and a significant decrease in PCEs. Additionally, the molecular symmetry of electron acceptor materials plays a critical role in determining photovoltaic performance.^28,29 For example, Facchetti and coworkers reported a series of electron acceptors, showing that asymmetric acceptors tend to hinder the formation of continuous end-group stacking.²⁹ This structural limitation results in lower charge carrier mobility, which further leads to slower hole transfer and increased exciton recombination in the BHJ blend. As a result, asymmetric electron acceptors generally exhibit significantly lower PCE compared to their symmetric counterparts. Moreover, the symmetry of electron-withdrawing units in polymers also affects the aggregation behavior of polymeric acceptors.^30,31 For instance, Yang and coworkers developed a series of polymer acceptors by combining asymmetric and axisymmetric electron-deficient building blocks.³¹ The polymer acceptor containing the most asymmetric structural unit produced the smallest domain size in the BHJ layer and yielded the lowest PCE. As the electron-deficient unit transitioned from asymmetric to axisymmetric, the domain size in the BHJ improved accordingly, leading to enhanced PCE. In contrast, non-fused ring electron acceptors (NFREAs) typically require fewer synthetic steps, resulting in lower synthesis costs.^32–35 Currently, the PCEs of NFREA-based OSCs also exceed 18%.^36–39 Therefore, investigating the molecular geometric configuration of NFREAs is crucial for the continued advancement of OSCs.

In this work, we designed and synthesized two NFREAs, namely TTCIC and TCIC (see Scheme 1), by incorporating 3,6-bis(octan-3-yloxy)thieno[3,2-b]thiophene (TT) and 3,4-bis(octan-3-yloxy)thiophene (T) into the conjugated backbone, which adopt S-shaped and C-shaped conformations, respectively. The introduction of oxygen atoms was designed to enable intramolecular S⋯O noncovalent interactions with adjacent cyclopentadithiophene (CDT) units, enhancing molecular planarity to improve charge carrier mobility, while the bulky side chains were expected to create substantial steric hindrance, facilitating favorable phase-separated morphologies. Compared with TTCIC, TCIC exhibits a wider bandgap, accompanied by an elevated LUMO energy level and a lowered HOMO energy level. When blended with the polymer donor PBDB-T, the two NFREAs exhibited dramatically different photovoltaic performances: the TTCIC-based OSC delivered a PCE of merely 0.60%, whereas the TCIC-based OSC achieved a significantly higher PCE of 8.66%. The origin of this performance difference was systematically investigated through aggregation behavior and BHJ morphology analysis.

Scheme 1 The synthetic routes of TTCIC and TCIC.

Results and discussion

Synthesis and characterization

The synthetic routes to TTCIC and TCIC are illustrated in Scheme 1. The intermediates 2,5-dibromo-3,6-bis(octan-3- yloxy)thieno[3,2-b]thiophene (2) and 2,5-dibromo-3,4-bis(octan-3-yloxy)thiophene (7) were prepared using low-cost commercial starting materials following previously reported procedures.^40,41 (4,4-Dioctyl-4H-cyclopenta[2,1-b:3,4-b′]dithien-2-yl)trimethylstannane (3) was prepared according to previously reported procedures.⁴² Subsequently, compounds 2 and 7 underwent Stille coupling reactions with (4,4-dioctyl-4H-cyclopenta[2,1-b:3,4-b′]dithiophen-2-yl)trimethylstannane, yielding compounds 4 and 8, respectively, which underwent Vilsmeier–Haack reactions to yield key dialdehyde intermediates 5 and 9. Finally, the target acceptors TTCIC and TCIC were obtained through Knoevenagel condensations. These novel intermediates and both acceptor materials with distinct configurations were characterized by ¹H nuclear magnetic resonance (NMR), ¹³C NMR, and mass spectrometry. Both acceptors exhibit good solubility in chloroform (>40 mg mL⁻¹). It should be noted that TTCIC exhibits strong aggregation properties in solution. When we tested its ¹³C NMR at concentrations of 20 mg mL⁻¹ and 40 mg mL⁻¹ in deuterated chloroform, respectively, only some of the carbon in the aromatic region was detected (Fig. S1 and S2). We also tested at a concentration of 40 mg mL⁻¹ in deuterated 1,1,2,2-tetrachloroethane at 100 °C (Fig. S3), again finding that only some of the aromatic carbon was detected. Finally, we verified its purity through elemental analysis. As shown in Fig. S4, both NFREAs exhibit good thermal stability, with an identical thermal decomposition temperature (5% weight loss) of 275 °C.

DFT calculations and optical, electrochemical and self-organization properties

To understand the geometries of both acceptors, density functional theory (DFT) calculations were performed using the B3LYP hybrid functional and the 6-31+G(d,p) basis set for ground-state geometry optimization.⁴³ Both acceptors were simplified by replacing their alkyl side chains with methyl groups. As shown in Fig. 1a, the distance between the oxygen atom on the TT unit and the sulfur atom in the adjacent CDT unit is 2.84 Å, significantly shorter than the sum of the van der Waals radii of O and S (3.32 Å). This indicates strong intramolecular S⋯O noncovalent interactions. Consequently, the dihedral angle between the TT unit and the adjacent CDT unit is 0°. In contrast, the distance between the oxygen atom on the T unit and the sulfur atom in the adjacent CDT unit increases to 2.92 Å, suggesting weaker intramolecular interactions compared to the TT-based acceptor. This increase is consistent with the larger dihedral angle (2.62°) observed between the T unit and the adjacent CDT unit. Collectively, the oxygen atoms on both the TT and T units contribute to highly planar conjugated backbones. Electron density distributions are shown in Fig. S5. At the LUMO level, electron density is delocalized across the entire conjugated backbone, whereas at the HOMO level, the acceptor end-groups exhibit diminished electron density. This suggests indicates that the two acceptors are more conducive to electron transfer. We further investigated the dipole moments of both acceptors.⁴⁴ The dipole moment change (Δμ_ge) between the ground state (μ_g) and excited state (μ_e) dipoles was calculated using time-dependent density functional theory (TD-DFT). A larger Δμ_ge in an organic molecule indicates a lower Coulomb binding energy, which facilitates exciton dissociation.^45,46 Δμ_ge was calculated according to the equation of Δμ_ge = [(μ_gx − μ_ex)² + (μ_gy − μ_ey)² + (μ_gz − μ_ez)²]^1/2. The corresponding values are listed in Table S1 (SI). Compared to TTCIC (Δμ_ge = 0 D), TCIC exhibited a slightly larger Δμ_ge of 0.247 D. This suggests that TCIC has a marginally weaker Coulomb binding energy, which is beneficial for exciton dissociation.

Fig. 1 The DFT calculated model of (a) TTCIC and TCIC with simplified side chains. (b) UV-Vis-NIR absorption spectra of TTCIC and TCIC in solution and films. (c) Energy level diagram of PBDB-T, TTCIC and TCIC. The GIWAXS images of (d) TTCIC and (e) TCIC. (f) The IGMH maps with intermolecular interaction energies with the dimer of TTCIC and TCIC.

The ultraviolet-visible-near infrared (UV-vis-NIR) absorption spectra of TTCIC and TCIC were recorded in dilute chlorobenzene solution (10⁻⁵ M) and thin films. The corresponding data are summarized in Table 1. As shown in Fig. 1b and Fig. S6, both acceptors exhibit similar absorption profiles in solution. TTCIC shows a maximum absorption wavelength (λ_max) at 801 nm with a molar extinction coefficient (ε) of 1.77 × 10⁵ M⁻¹ cm⁻¹, while TCIC exhibits a blue-shifted λ_max at 790 nm with ε = 2.12 × 10⁵ M⁻¹ cm⁻¹. This blue shift in TCIC suggests weaker intramolecular charge transfer compared to TTCIC. In thin films, both acceptors display markedly red-shifted and broadened absorption spectra compared to their solution states, attributed to intermolecular interactions in the solid state. The λ_max of TTCIC in the film is red-shifted by 125 nm to 926 nm, whereas TCIC shows an 88 nm red shift to 878 nm. This indicates stronger intermolecular charge transfer in TTCIC, likely resulting from its more planar conjugated backbone. The optical parameters are calculated to be 1.18 eV and 1.25 eV, respectively. These results indicate that incorporating TT and T units into the conjugated skeleton to regulate molecular geometry has a significant effect on the absorption properties of NFREAs.

Table 1 Optical and electrochemical parameters of TTCIC and TCIC

Acceptor	λmax,solution (nm)	λmax,film (nm)	E^optg (eV)	HOMO (eV)	LUMO (eV)
TTCIC	801	926	1.18	−5.35	−3.98
TCIC	790	878	1.25	−5.40	−3.89

---

Cyclic voltammetry was performed to evaluate energy levels, with the polymer donor PBDB-T measured under identical conditions for comparison. As shown in Fig. 1c and Fig. S7, TTCIC exhibits HOMO and LUMO levels of −5.35 eV and −3.98 eV, respectively. When the central core of the TT unit is replaced by the T unit, TCIC exhibits a lowered HOMO level (−5.40 eV) and a raised LUMO level (−3.89 eV). The electrochemical bandgaps for TTCIC and TCIC are 1.37 eV and 1.51 eV, respectively, which are 0.19 eV and 0.26 eV larger than their corresponding E^opt_gs. This difference between the optically and electrochemically determined bandgaps can be attributed to the exciton binding energy in organic semiconductors.^47,48 As shown in Fig. S7 and S8, PBDB-T exhibits HOMO/LUMO levels of −5.32/−3.51 eV and a λ_max of 629 nm in thin films, respectively. Both acceptors show well-matched energy levels and complementary absorption with the PBDB-T donor polymer (Fig. S8), indicating their good potential for application in OSCs.

The microstructures of pure acceptor films were investigated using grazing-incidence wide-angle X-ray scattering (GIWAXS),^49,50 as shown in Fig. 1d and e. Both acceptors exhibit strong (010) diffraction peaks at 1.79 Å⁻¹ and 1.78 Å⁻¹ in the in-plane (IP) direction, respectively. These peaks are attributed to π–π stacking of the acceptors, with corresponding distances of 3.51 Å and 3.53 Å for TTCIC and TCIC, respectively. Additionally, the crystal coherence lengths (CCL) for the (010) diffraction peaks were also calculated to be 31.3 Å and 26.3 Å for TTCIC and TCIC, respectively. Additionally, we further quantified the intermolecular interaction energies of acceptor dimers based on Hirshfeld partition (IGMH) mapped distinct interaction regions.⁵¹ As shown in Fig. 1f, TTCIC exhibits a stronger intermolecular interaction energy of −20.06 kcal mol⁻¹ than TCIC (−19.92 kcal mol⁻¹). The smaller π–π stacking distance and stronger intermolecular interaction to form longer CCL in TTCIC correlates directly with its more pronounced red shift in film absorption. The stronger intermolecular interaction observed in TTCIC compared to TCIC is likely attributed to its enhanced molecular planarity and extended π-conjugation.

Photovoltaic properties

To evaluate the photovoltaic properties of both electron acceptors, we selected the polymer donor PBDB-T, which exhibits well-matched energy levels and complementary absorption (Fig. 2a). Devices were fabricated with a conventional architecture: ITO/PEDOT:PSS/PBDB- T:acceptor/PDINO/Al. The optimal current density–voltage (J–V) characteristics and external quantum efficiency (EQE) spectra are presented in Fig. 2b and c, with corresponding photovoltaic parameters listed in Table 2. Surprisingly, PBDB-T:TTCIC-based OSCs demonstrated a very low PCE of 0.60%, with an open-circuit voltage (V_OC) of 0.648 V, a short-circuit current density (J_SC) of 1.57 mA cm⁻², and a fill factor (FF) of 59.5%. When the central core was changed from TT to T, PBDB-T:TCIC-based devices achieved a significantly higher PCE of 8.66% with comprehensively enhanced photovoltaic parameters: a V_OC of 0.774 V, a J_SC of 17.30 mA cm⁻², and a FF of 64.7%. Compared to TTCIC-based devices, the increased V_OC in PBDB-T:TCIC-based OSCs is attributed to the elevated LUMO level of TCIC, while the improved J_SC and FF might be attributed to the optimized BHJ morphology. The EQE spectra of both optimized devices span an identical 300–1000 nm range, mirroring their corresponding absorption profiles. Photocurrent generation in the 300–700 nm window predominantly originates from the PBDB-T donor, while contributions beyond 700 nm stem exclusively from the TTCIC/TCIC acceptors. This distinct spectral assignment confirms that the substantially enhanced J_SC in PBDB-T:TCIC-based devices arises primarily from better BHJ morphology. The calculated integrated current densities of 1.51 mA cm⁻² (PBDB-T:TTCIC) and 16.90 mA cm⁻² (PBDB-T:TCIC) exhibit a <5% deviation from J–V measurements.

Fig. 2 (a) Chemical structure of PBDB-T. (b) J–V and (c) EQE curves of the optimized PBDB-T:TTCIC and PBDB-T:TCIC-based OSCs. The J_ph–V_eff (d), J_SC–P_light (e) and V_OC–P_light (f) curves of PBDB-T:TTCIC- and PBDB-T:TCIC-based OSCs.

Table 2 Photovoltaic parameters of PBDB-T:TTCIC- and PBDB-T:TCIC-based OSCs under AM1.5 G (100 mW cm⁻²) illumination^a

Active layer	VOC (V)	JSC (mA cm^−2)	FF (%)	PCE (%)
a Average values with standard deviations are obtained from 10 independent cells.
PBDB-T:TTCIC	0.648 (0.644 ± 0.003)	1.57 (1.54 ± 0.050)	59.5 (59.6 ± 0.28)	0.60 (0.58 ± 0.02)
PBDB-T:TCIC	0.774 (0.775 ± 0.001)	17.30 (16.30 ± 0.663)	64.7 (65.2 ± 0.316)	8.66 (8.24 ± 0.11)

---

Charge generation, transport and recombination

To investigate exciton dissociation and charge collection in the devices, we measured the photocurrent density (J_ph) as a function of the effective voltage (V_eff). Here, J_ph = J_L − J_D, where J_L and J_D are the current densities under illumination and in the dark, respectively. V_eff = V₀ − V, where V is the applied voltage and V₀ is the voltage when J_ph is zero. Fig. 2d shows that the exciton dissociation probability (P_diss) and charge collection probability (P_coll) were calculated based on the ratio of the J_ph/saturated photocurrent density (J_sat). The PBDB-T:TTCIC and PBDB-T:TCIC devices exhibited P_diss/P_coll values of 67.5/50.2% and 87.4/72.0%, respectively. The electrostatic potential (ESP) of these materials was also investigated as shown in Fig. S9. The overall average ESP values were calculated to be −63.58, 198.14, and 202.80 meV for PBDB-T, TTCIC, and TCIC, respectively. The large ESP offset (266.38 vs. 261.71 meV) in PBDB-T:TCIC-based OSCs is beneficial for exciton dissociation. The lower exciton dissociation efficiency and inferior charge collection capability in PBDB-T:TTCIC-based OSCs reduced its EQE response, resulting in diminished J_SC.

Charge transport properties were evaluated via the space-charge-limited current (SCLC) method. The electron-only devices were fabricated with the structure of ITO/TIPD/active layer/PDINO/Al, and hole-only devices were fabricated by ITO/PEDOT:PSS/active layer/Au. As depicted in Fig. S10, first, we investigated the electron mobility (μ_e) of the two acceptors. TCIC exhibited an electron mobility of 9.61 × 10⁻⁵ cm² V⁻¹ s⁻¹, which is more than twice that of TTCIC (4.53 × 10⁻⁵ cm² V⁻¹ s⁻¹). This suggests that the highly ordered edge-on aromatic packing structure in TTCIC contributes weakly to vertical electron transport. After blending with the polymer, the μ_e and hole (μ_h) mobilities of BDB-T:TTCIC films were measured to be 7.90 × 10⁻⁵ and 2.34 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. In contrast, the PBDB-T:TCIC films were determined to be 1.15 × 10⁻⁴ and 1.96 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. Their corresponding electron/hole mobility (μ_h/μ_e) ratios were calculated to be 3.0, and 1.7 for PBDB-T:TTCIC and PBDB-T:TCIC films, respectively. The more balanced charge transport of the PBDB-T:TCIC blend is beneficial for reducing charge recombination, thus yielding increased J_SC and FF.

Furthermore, to elucidate the charge recombination dynamics in both OSCs, we systematically measured the J_SC as a function of light intensities (P_light) as presented in Fig. 2e.^52,53 The photocurrent response follows a power-law dependence expressed as J_SC ∝ P^S_light. The exponential factor S serves as a critical indicator of recombination kinetics. When the S value approaches unity, it signifies minimal bimolecular charge recombination within the device. The S values were determined to be 0.961, and 0.994 for PBDB-T:TTCIC- and PBDB-T:TCIC-based devices, respectively. This marked divergence demonstrates significantly suppressed bimolecular recombination in PBDB-T:TCIC systems compared to their PBDB-T:TTCIC counterparts, hereby explaining the enhanced J_SC and FF in the PBDB-T:TCIC-based OSCs. Additionally, the relationship between V_OC and P_light was also investigated (Fig. 2f). It is described as V_OC ∝ nkT/q ln(P_light). Usually, a larger n value represents a more serious trap-assisted recombination in devices. The calculated n values were 1.24 and 1.11 for PBDB-T:TTCIC- and PBDB-T:TCIC-based devices, respectively, which indicated that trap-assisted recombination was suppressed effectively in the PBDB-T:TCIC-based OSC.

To deeply elucidate hole transport dynamics in blend films, femtosecond transient absorption (fs-TA) spectroscopy was employed to probe hole transfer processes as shown in Fig. 3.⁵⁴ Considering that the ground-state bleaching (GSB) signal peaks of organic semiconductors typically coincide with the peaks of their steady-state absorption spectra. Taking the PBDB-T neat film as an example, the PBDB-T film exhibits absorption peaks from 500 nm to 650 nm, and the corresponding GSB signals are observed in the same spectral range.⁵⁵ Given that absorption at 800 nm originates exclusively from TTCIC or TCIC, an excitation wavelength of 800 nm was chosen to selectively excite the acceptor. This eliminates interference from the PBDB-T and enables investigation of hole transfer from the acceptor to the donor. Fig. 3b and e present the corresponding time-resolved absorption spectra at specified delay intervals. After light excitation, GSB signatures immediately appears at approximately 750 nm, which can be attributed to the acceptor materials due to their similarity to the corresponding steady-state absorption spectra. Subsequently, the GSB signals of the acceptors gradually diminish, and after 1600 ps, the GSB signatures of both TTCIC and TCIC almost completely disappeared. During the period from 0.2 ps to 100 ps, a continuous rise in absorption at 580 nm and 630 nm is observed in Fig. 3e, which is attributed to the GSB signature of PBDB-T in PBDB-T:TCIC films.⁵⁵ In contrast, the GSB signal of PBDB-T in PBDB-T:TTCIC films remains very weak (Fig. 3b). Since the singlet energies of both low bandgap acceptors are lower than that of the large bandgap polymer donor, it is impossible to occur singlet energy transfer from acceptors to PBBD-T. Consequently, the rapid onset of PBDB-T GSB should originate from hole transfer from photoexcited acceptors to adjacent PBDB-T. Notably, the substantially weaker GSB signals of PBDB-T in PBDB-T:TTCIC films indicates inefficient hole transfer from TTCIC to PBDB-T. Kinetic traces at 630 nm for both blend films were quantitatively analyzed (Fig. 4c and f). Both systems exhibit rapid decay components, with the PBDB-T:TCIC blend demonstrating accelerated hole transfer with a time of 0.46 ps, compared with 0.64 ps for PBDB-T:TTCIC. The accelerated hole transport in the PBDB-T:TCIC system indicates more efficient charge separation, which supports the higher J_SC observed in PBDB-T:TCIC-based OSCs.

Fig. 3 2D colour transient absorption spectra of (a) PBDB-T:TTCIC and (d) PBDB-T:TCIC films. The representative spectra at indicated delay times of (b) PBDB-T:TTCIC and (e) PBDB-T:TCIC films under 800 nm excitation. (c) TA kinetics and fitting curves at 630 nm of (c) PBDB-T:TTCIC and (f) PBDB-T:TCIC films.

Fig. 4 The AFM (a) high and (b) phase images and (c) GIWAXS images of PBDB-T:TTCIC films. The AFM (d) high and (e) phase images and (f) GIWAXS images of PBDB-T:TCIC films. (g) The IP and OOP extracted line-cut profiles of PBDB-T:TTCIC and PBDB-T:TCIC blend films.

Blend morphologies and microstructures

To gain a deeper understanding of different J_SCs and FFs of both OSCs, atomic force microscopy (AFM) was utilized to characterize blend film surface morphologies (Fig. 4). The PBDB-T:TTCIC blend exhibits a markedly elevated root-mean-square surface roughness (R_q) of 5.40 nm, arising from its substantial crystalline domains, which could be attributed to the strong crystallization of TTCIC. Conversely, the PBDB-T:TCIC film demonstrates a smoother surface with an R_q of 1.26 nm. The large crystallization size in the PBDB-T:TTCIC film is not beneficial for exciton dissociation and transportation, yielding low J_SC and FF.^56–58 Subsequently, GIWAXS measurements were conducted to probe molecular packing arrangements. Two-dimensional GIWAXS patterns and corresponding one-dimensional profiles are presented in Fig. 4c, f and g. Both blends exhibit (010) diffraction peaks at 1.79 Å⁻¹ for the PBDB-T:TTCIC film and 1.78 Å⁻¹ for the PBDB-T:TCIC film, corresponding to π–π stacking distances of 3.51 Å and 3.53 Å, respectively, along both OOP and IP directions. These stacking parameters align closely with those observed in pristine acceptor films. The PBDB-T:TTCIC film displays intensified (010) diffraction intensity in the IP direction relative to the OOP direction, indicating a preferential edge-on molecular orientation. This suboptimal alignment impedes efficient charge carrier transport and extraction within the OSC architecture, consequently yielding an anomalously low J_SC and diminished PCE. In contrast, the PBDB-T:TCIC blend exhibits stronger (010) diffraction in the OOP direction compared to the IP direction, signifying dominant face-on stacking. It should be noted that the TCIC pure film adopts the edge-on orientation, while the PBDB-T:TCIC blend film exhibits the face-on orientation. GIWAXS measurements confirm that the PBDB-T polymer itself adopts a predominantly face-on orientation (Fig. S11). This orientation transition may be attributed to the fact that PBDB-T tends to precipitate first during the film formation process, thereby influencing the molecular packing of TCIC through intermolecular interactions, and promoting its face-on alignment in the blend.⁵⁹ This favorable orientation promotes effective vertical charge transport, enabling balanced J_SC and PCE values. Collectively, the severely compromised photovoltaic performance of PBDB-T:TTCIC is primarily due to its detrimental edge-on orientation, which increases charge carrier recombination losses.

Conclusions

In summary, we report two NFEAs, namely TTCIC (S-shaped) and TCIC (C-shaped). Both NFEAs exhibit enhanced molecular planarity enabled by intramolecular S⋯O noncovalent locking. Upon substituting the TT core unit with a T moiety, diminished intermolecular interactions occur, resulting in a blue-shifted absorption spectrum, downshifted HOMO energy level (−5.40 eV vs. −5.35 eV), and uplifted LUMO energy level (−3.89 eV vs. −3.98 eV). The S-shaped TTCIC exhibited a μ_ge of 0 D. In contrast, C-shaped TCIC demonstrated a significantly enhanced μ_ge of 0.217 D, indicating good exciton dissociation capability. Consistent with this, ESP analysis reveals a larger offset in PBDB-T:TCIC blends (266.38 meV) than that of PBDB-T:TTCIC (261.71 meV), further confirming facilitated exciton dissociation. Additionally, edge-on stacking of the PBDB-T:TTCIC film, coupled with larger phase separation sizes, results in highly inefficient hole transfer from the acceptor to the polymer donor. By contrast, face-on stacking of the PBDB-T:TCIC film produces moderate hole transfer. As a result, PBDB-T:TTCIC-based OSCs deliver a very low PCE of 0.60% with a poor J_SC of 1.57 mA cm⁻² and a FF of 59.5%. In contrast, PBDB-T:TCIC-based OSCs achieved a significantly improved PCE of 8.66%, with a better J_SC of 17.30 mA cm⁻² and a FF of 64.7%. These results demonstrate that fine-tuning of NFREA geometric configurations is essential for optimizing the molecular stacking orientation and enhancing the OSC performance.

Conflicts of interest

The authors declare no competing interests.

Data availability

The data supporting this article have been included as part of the SI. The Supplementary Information file contains additional data and figures that support the findings of this study, including instruments and measurements, synthetic details, TGA, DFT, CV, absorption, SCLC, and GIWAXs images. See DOI: https://doi.org/10.1039/d5tc02718g.

Acknowledgements

The authors acknowledge the funding support from the National Natural Science Foundation of China (NSFC 22475136, 22433005 and 22173062). The authors gratefully acknowledge the cooperation of the beamline scientists at the BSRF-1W1A beamline for the two-dimensional GIWAXS measurements.

Notes and references

1. N. S. Sariciftci, L. Smilowitz, A. J. Heeger and F. Wudl, Science, 1992, 258, 1474–1476.
2. G. Yu, J. Gao, J. C. Hummelen, F. Wudl and A. J. Heeger, Science, 1995, 270, 1789–1791.
3. Y. Liu, B. Liu, C.-Q. Ma, F. Huang, G. Feng, H. Chen, J. Hou, L. Yan, Q. Wei, Q. Luo, Q. Bao, W. Ma, W. Liu, W. Li, X. Wan, X. Hu, Y. Han, Y. Li, Y. Zhou, Y. Zou, Y. Chen, Y. Li, Y. Chen, Z. Tang, Z. Hu, Z.-G. Zhang and Z. Bo, Sci. China: Chem., 2022, 65, 224–268.
4. Y. Liu, B. Liu, C.-Q. Ma, F. Huang, G. Feng, H. Chen, J. Hou, L. Yan, Q. Wei, Q. Luo, Q. Bao, W. Ma, W. Liu, W. Li, X. Wan, X. Hu, Y. Han, Y. Li, Y. Zhou, Y. Zou, Y. Chen, Y. Liu, L. Meng, Y. Li, Y. Chen, Z. Tang, Z. Hu, Z.-G. Zhang and Z. Bo, Sci. China: Chem., 2022, 65, 1457–1497.
5. C. An and J. Hou, Acc. Mater. Res., 2022, 3, 540–551.
6. G. Wang, M. A. Adil, J. Zhang and Z. Wei, Adv. Mater., 2018, 31, 1805089.
7. Y. Lin, J. Wang, Z.-G. Zhang, H. Bai, Y. Li, D. Zhu and X. Zhan, Adv. Mater., 2015, 27, 1170–1174.
8. J. Yuan, Y. Zhang, L. Zhou, G. Zhang, H.-L. Yip, T.-K. Lau, X. Lu, C. Zhu, H. Peng, P. A. Johnson, M. Leclerc, Y. Cao, J. Ulanski, Y. Li and Y. Zou, Joule, 2019, 3, 1140–1151.
9. J. Wang, P. Xue, Y. Jiang, Y. Huo and X. Zhan, Nat. Rev. Chem., 2022, 6, 614–634.
10. C. Yan, S. Barlow, Z. Wang, H. Yan, A. K. Y. Jen, S. R. Marder and X. Zhan, Nat. Rev. Mater., 2018, 3, 18003.
11. W. Zhao, D. Qian, S. Zhang, S. Li, O. Inganäs, F. Gao and J. Hou, Adv. Mater., 2016, 28, 4734–4739.
12. K. Li, Q. Xie, C. An, J. Zhong, Y. Yao, C. Hong, Y. Chen, H. Geng, Y. Wu, W. Zhang, Q. Liao and H. Fu, Angew. Chem., Int. Ed., 2025, 64, e202506415.
13. Y. Jiang, S. Sun, R. Xu, F. Liu, X. Miao, G. Ran, K. Liu, Y. Yi, W. Zhang and X. Zhu, Nat. Energy, 2024, 9, 975–986.
14. J. Song, C. Zhang, C. Li, J. Qiao, J. Yu, J. Gao, X. Wang, X. Hao, Z. Tang, G. Lu, R. Yang, H. Yan and Y. Sun, Angew. Chem., Int. Ed., 2025, 63, e202404297.
15. L. Zhu, M. Zhang, G. Zhou, Z. Wang, W. Zhong, J. Zhuang, Z. Zhou, X. Gao, L. Kan, B. Hao, F. Han, R. Zeng, X. Xue, S. Xu, H. Jing, B. Xiao, H. Zhu, Y. Zhang and F. Liu, Joule, 2024, 8, 3153–3168.
16. R. Ma, B. Zou, Y. Hai, Y. Luo, Z. Luo, J. Wu, H. Yan and G. Li, Adv. Mater., 2025, 37, 2500861.
17. Z. Ling, J. Wu, J. P. Jurado, C. E. Petoukhoff, S. Y. Jeong, D. Naphade, M. Babics, X. Chang, H. Faber, S. Doukas, E. Lidorikis, M. I. Nugraha, M. He, M. Alqurashi, Y. Lin, X. Sun, H. Hu, H. Y. Woo, S. De Wolf, L. Tsetseris, F. Laquai, D. Yu, E. Wang and T. D. Anthopoulos, Mater. Sci. Eng. R Rep., 2025, 163, 100922.
18. Y. Sun, L. Wang, C. Guo, J. Xiao, C. Liu, C. Chen, W. Xia, Z. Gan, J. Cheng, J. Zhou, Z. Chen, J. Zhou, D. Liu, T. Wang and W. Li, J. Am. Chem. Soc., 2024, 146, 12011–12019.
19. S. Guan, Y. Li, C. Xu, N. Yin, C. Xu, C. Wang, M. Wang, Y. Xu, Q. Chen, D. Wang, L. Zuo and H. Chen, Adv. Mater., 2024, 36, 2400342.
20. X. Zhu, C. Gu, Y. Cheng, H. Lu, X. Wang, G. Ran, W. Zhang, Z. Tang, Z. Bo and Y. Liu, Adv. Mater., 2025, 37, 2507529.
21. H. Mou, Y. Yin, H. Chen, J. Xu, J. Ding, C. Ju, J. Zhu, Y. Wang, W. Chen, G. Xu, T. Zhang, J. Li, Y. Li and Y. Li, J. Am. Chem. Soc., 2025, 147, 21241–21251.
22. P. Mao, B. Zhang, H. Zhang, M. Jiang, Y. Wang, Y. Li, J. Wu, J.-L. Wang and Q. An, Mater. Sci. Eng. R Rep., 2025, 165, 101022.
23. Y. Jiang, K. Liu, F. Liu, G. Ran, M. Wang, T. Zhang, R. Xu, H. Liu, W. Zhang, Z. Wei, Y. Cui, X. Lu, J. Hou and X. Zhu, Adv. Mater., 2025, 37, 2500282.
24. J. Lee, R. Singh, D. H. Sin, H. G. Kim, K. C. Song and K. Cho, Adv. Mater., 2015, 28, 69–76.
25. Y. Bai, L. Hong, Y. Dou, S. Zhu, H. Tang, D. Liu, Y. Cao, J. Chen, S. Chen, L. Shao, Z. Hu, D. Tang, K. Zhang, S.-J. Su, C. Liu and F. Huang, ACS Energy Lett., 2024, 9, 1786–1795.
26. Z. Xia, J. Zhang, X. Gao, W. Song, J. Ge, L. Xie, X. Zhang, Z. Liu and Z. Ge, ACS Appl. Mater. Interfaces, 2021, 13, 23983–23992.
27. Q. Guo, R. Ma, J. Hu, Z. Wang, H. Sun, X. Dong, Z. Luo, T. Liu, X. Guo, X. Guo, H. Yan, F. Liu and M. Zhang, Adv. Funct. Mater., 2020, 30, 2000383.
28. Y. Chen, G. Wang, L. Yang, J. Wu, F. S. Melkonyan, Y. Huang, Z. Lu, T. J. Marks and A. Facchetti, J. Mater. Chem. C, 2018, 6, 847–854.
29. G. Li, F. Qin, R. M. Jacobberger, S. Mukherjee, L. O. Jones, R. M. Young, R. M. Pankow, B. P. Kerwin, L. Q. Flagg, D. Zheng, L.-W. Feng, K. L. Kohlstedt, V. K. Sangwan, M. C. Hersam, G. C. Schatz, D. M. DeLongchamp, M. R. Wasielewski, Y. Zhou, A. Facchetti and T. J. Marks, Joule, 2023, 7, 2152–2173.
30. J. Wu, C. Fu, R. M. Pankow, Y. Chen, D. Zheng, Z. Lu, Y. Huang, T. J. Marks and A. Facchetti, Chem. Mater., 2023, 35, 10106–10118.
31. G. Park, Y. Cho, S. Lee, S. Kim, K. C. Lee and C. Yang, Phys. Chem. Chem. Phys., 2023, 25, 17001–17009.
32. P. Jiang, Y. Liu, J. Song and Z. Bo, Acc. Chem. Res., 2024, 57, 3419–3432.
33. Q. Shen, C. He, S. Li, L. Zuo, M. Shi and H. Chen, Acc. Mater. Res., 2022, 3, 644–657.
34. X. Liu, K. Li, Z. Mei, Y. Chen, Q. Xie, J. Li, Z. Lu, Y. Wu, Q. Liao, C. An and H. Fu, Solar RRL, 2024, 8, 2400153.
35. K. Li, X. Liu, J. Zhong, Y. Chen, W. Zhang, P. Wang, Y. Wu, Q. Liao, C. An and H. Fu, Small, 2024, 20, 2405525.
36. X. Gu, R. Zeng, Y. Hou, N. Yu, J. Qiao, H. Li, Y. Wei, T. He, J. Zhu, J. Deng, S. Tan, C. E. Zhang, Y. Cai, G. Long, X. Hao, Z. Tang, F. Liu, X. Zhang and H. Huang, Angew. Chem., Int. Ed., 2024, 63, e202407355.
37. Z. Bo, X. Zheng, X. Jiang, W. Liu, C. Cao, N. Wei, Z. Wei, L. Zhang, H. Wu, A. Zhang, C. Li, Y. Liu, H. Lu, X. Xu and Z. Tang, Angew. Chem., Int. Ed., 2024, 63, e202412854.
38. R. Zeng, M. Zhang, X. Wang, L. Zhu, B. Hao, W. Zhong, G. Zhou, J. Deng, S. Tan, J. Zhuang, F. Han, A. Zhang, Z. Zhou, X. Xue, S. Xu, J. Xu, Y. Liu, H. Lu, X. Wu, C. Wang, Z. Fink, T. P. Russell, H. Jing, Y. Zhang, Z. Bo and F. Liu, Nat. Energy, 2024, 9, 1117–1128.
39. Y. Wang, M. Yang, Z. Chen, J. Zhong, F. Zhao, W. Wei, X. Yuan, W. Zhang, Z. Ma, Z. He, Z. Liu, F. Huang, Y. Cao and C. Duan, Nat. Commun., 2025, 16, 5449.
40. Q. Xie, Z. Mei, K. Li, Y. Chen, J. Li, X. Liu, Y. Wu, Q. Liao, C. An and H. Fu, Dyes Pigm., 2024, 231, 112415.
41. J. Li, K. Li, Z. Mei, Y. Chen, Q. Xie, C. Yu, X. Liu, Y. Wang, Y. Wu, Q. Liao, C. An and H. Fu, J. Energy Chem., 2024, 96, 501–508.
42. F. Lincker, N. Delbosc, S. Bailly, R. De Bettignies, M. Billon, A. Pron and R. Demadrille, Adv. Funct. Mater., 2008, 18, 3444–3453.
43. M. J. Frisch, et al., Gaussian 03, Revision B.04, Gaussian, Inc., Pittsburgh PA, 2003.
44. T. Zhang, C. An, Y. Xu, P. Bi, Z. Chen, J. Wang, N. Yang, Y. Yang, B. Xu, H. Yao, X. Hao, S. Zhang and J. Hou, Adv. Mater., 2022, 34, 2207009.
45. B. Carsten, J. M. Szarko, H. J. Son, W. Wang, L. Lu, F. He, B. S. Rolczynski, S. J. Lou, L. X. Chen and L. Yu, J. Am. Chem. Soc., 2011, 133, 20468–20475.
46. E. Collado-Fregoso, P. Boufflet, Z. Fei, E. Gann, S. Ashraf, Z. Li, C. R. McNeill, J. R. Durrant and M. Heeney, Chem. Mater., 2015, 27, 7934–7944.
47. C. An, S. R. Puniredd, X. Guo, T. Stelzig, Y. Zhao, W. Pisula and M. Baumgarten, Macromolecules, 2014, 47, 979–986.
48. J.-L. Bredas, Mater. Horiz., 2014, 1, 17–19.
49. H. Alexander, B. Wim, G. James, S. Eric, G. Eliot, K. Rick, M. Alastair, C. Matthew, R. Bruce and P. Howard, J. Phys.: Conf. Ser., 2010, 247, 012007.
50. E. Gann, A. T. Young, B. A. Collins, H. Yan, J. Nasiatka, H. A. Padmore, H. Ade, A. Hexemer and C. Wang, Rev. Sci. Instrum., 2012, 83, 045110.
51. B. Fan, H. Gao, L. Yu, R. Li, L. Wang, W. Zhong, Y. Wang, W. Jiang, H. Fu, T. Chen, B. Kan, S.-W. Tsang and A. K.-Y. Jen, Angew. Chem., Int. Ed., 2025, 64, e202418439.
52. S. R. Cowan, A. Roy and A. J. Heeger, Phys. Rev. B:Condens. Matter Mater. Phys., 2010, 82, 245207.
53. I. Riedel, J. Parisi, V. Dyakonov, L. Lutsen, D. Vanderzande and J. C. Hummelen, Adv. Funct. Mater., 2004, 14, 38–44.
54. H. Lu, H. Wang, G. Ran, W. Liu, H. Huang, X. Jiang, Y. Liu, X. Xu and Z. Bo, CCS Chem., 2024, 6, 1757–1766.
55. R. Jasiunas, H. Zhang, J. Yuan, X. Zhou, D. Qian, Y. Zou, A. Devižis, J. Šulskus, F. Gao and V. Gulbinas, J. Phys. Chem. C, 2020, 124, 21283–21292.
56. K. Ma, C. An, T. Zhang, Q. Lv, S. Zhang, P. Zhou, J. Zhang and J. Hou, J. Mater. Chem. C, 2020, 8, 2301–2306.
57. P. Wan, C. An, T. Zhang, K. Ma, N. Liang, Y. Xu, S. Zhang, B. Xu, J. Zhang and J. Hou, Polym. Chem., 2020, 11, 1629–1636.
58. H. Hu, K. Jiang, P. C. Y. Chow, L. Ye, G. Zhang, Z. Li, J. H. Carpenter, H. Ade and H. Yan, Adv. Energy Mater., 2018, 8, 1701674.
59. Y. Zhou, M. Li, N. Yu, S. Shen, J. Song, Z. Ma and Z. Bo, ACS Appl. Mater. Interfaces, 2022, 14, 6039–6047.

---