A bithiophene imide-based polymer donor for alloy-like ternary organic solar cells with over 20.5% efficiency and enhanced stability

Abstract

Ternary organic solar cells (TOSCs) based on dual polymer donors offer enhanced absorption and stability by broadening spectral coverage and refining phase morphology. However, the inherent chain entanglement of dual polymer donors leads to sizable steric hindrance, hindering efficient mixing and posing challenges for further performance improvements. Here, we introduce a new polymer donor PBTI-FR, featuring a bithiophene imide (BTI) acceptor unit, which is specifically tailored to form a dual-polymer-donor TOSC with PM6 and L8-BO. Polymer donor PBTI-FR exhibits strong dipole moments and favorable miscibility with another polymer donor PM6, promoting a stable alloy donor structure. This alloy donor strategy not only reduces energy loss but also strengthens intermolecular interactions and fine-tunes film nanomorphology. Consequently, exciton dissociation and charge transport are improved, delivering a remarkable power conversion efficiency of 20.52%, among the highest values reported for OSCs, alongside an exceptional fill factor of 82.55%. Furthermore, the ternary devices exhibit excellent thermal stability, retaining over 92.2% of their initial performance after 1008 h of heating, underscoring the effectiveness of the dual-polymer-donor alloy design in countering performance degradation. This work highlights a versatile route for high-performance OSCs through the synergistic design of alloy donors with well-aligned energy levels and precisely tuned film morphologies, enabling both superior efficiency and stability.

---

Broader context

Most organic solar cell (OSC) research has centered on binary systems, yet challenges such as open-circuit voltage loss and suboptimal active layer morphology persist. Ternary organic solar cells have emerged as an effective strategy to address these issues. While polymer donors typically offer superior stability, tunable crystallinity, and broader visible light absorption compared to non-fullerene small-molecule acceptors, their role as the third component in ternary systems has been limited due to the steric hindrance that hampers efficient energy transfer. Here, we introduce a new polymer donor, PBTI-FR, based on bithiophene imide (BTI), into the PM6:L8-BO system to form an alloy-like structure with PM6. PBTI-FR exhibits lower recombination energy, modulating the intermolecular interactions within the active layer. This leads to reduced energy loss, promotes the formation of a nanofibrous phase-separated morphology, and lowers the defect density within the active layer. As a result, the PM6:PBTI-FR:L8-BO ternary device achieved an impressive power conversion efficiency of 20.52%, accompanied by outstanding thermal stability, maintaining over 92.2% of their initial performance after 1008 h heating. This work underscores the critical role of the ternary strategy in minimizing energy loss and fine-tuning the active layer morphology to achieve high-performance OSCs.

Introduction

Organic solar cells (OSCs) have garnered significant attention in both academia and industry due to their light weight, mechanical flexibility, cost-effectiveness, and potential for large-area and high throughput manufacturing.^1–4 Over the past decade, advancements in molecular design^5–13 and device engineering^14–18 have propelled OSCs toward higher power conversion efficiencies (PCEs). Despite single-junction OSCs having achieved efficiencies exceeding 20%,^19–23 their performance still falls far short of the theoretical Shockley–Queisser (SQ) limit. Major challenges for binary systems, including the limited absorption range of active layers, energy level mismatches between donors and acceptors, and non-radiative recombination losses, continue to hinder photon harvesting and result in substantial voltage loss.^20,24–29 Moreover, a fundamental trade-off in the binary system exists between the phase co-mingling (which promotes exciton dissociation) and phase separation (which facilitates charge transport). Undesired phase separation can exacerbate carrier recombination, lowering both the open-circuit voltage (V_oc) and the fill factor (FF),^21,30–33 thereby constraining the overall photovoltaic performance of OSCs.

Introducing a third component into binary systems to form ternary OSCs (TOSCs) has proven to be an effective strategy to overcome these limitations. The ternary strategy enables fine modulation of energy levels, crystallinity, and phase morphology, thus boosting charge transport and reducing recombination losses.^34–40 Among various third-component candidates, non-fullerene small-molecule acceptors (NFAs) are widely utilized due to their ready availability and good compatibility with donor and acceptor materials. However, NFAs are prone to over-aggregation and degradation under thermal stress, complicating active-layer morphology control and adversely affecting long-term device stability.^41–44 Additionally, their strong absorption in the infrared region, mismatched with the solar spectrum peak (500–600 nm), limits the spectral response breadth and reduces light harvesting efficiency.

To address these challenges, incorporating a polymer donor (P_D) as the third component offers pronounced advantages, including improved morphology control, more complementary absorption in the visible range, and enhanced device stability. The mechanisms underpinning this enhancement include charge transfer, alloy formation, energy transfer, and parallel models.⁴⁵ For instance, high-efficiency TOSCs using PM6 and D18 as polymer donors have demonstrated the ability to form a fine-tuned dual-fiber network within the active layer,^46,47 and minimize exciton recombination losses and increase output power. Similarly, adding the broad bandgap polymer donor PTBzBI-dF to the D18-Cl:Y6 system⁴⁸ not only broadens the light-absorbing range but also establishes an ideal interlevel energy configuration that facilitates charge separation and transport while improving thermal stability.⁴⁹ Despite these advances, the inherent entanglement of long polymer chains, associated with their complex conformational characters, results in sizable steric hindrance when P_Ds are introduced as the third component into the host system.⁵⁰ This steric hindrance effect can disrupt efficient energy transfer pathways, compromising the performance of dual-polymer-donor-based TOSCs.⁵¹ Consequently, highly efficient TOSCs with dual-polymer donors remain rare. Achieving optimal energy level alignment and fine-tuned morphology through the design of highly compatible P_Ds is thus critical for further TOSC performance improvement.

In this work, we designed and synthesized a polymer donor, named PBTI-FR based on the bithiophene imide (BTI) acceptor unit,⁹ tailored explicitly for constructing high-performance TOSCs with dual polymer donors. Compared with conventional acceptor units such as BDD and DTBT,^5,52 BTI features a larger dipole moment and superior solubility, thereby enhancing interchain interactions and improving processability. Moreover, BTI demonstrates strong electron-withdrawing capability, deep highest occupied molecular orbital (HOMO) energy level, and excellent backbone planarity—well-desired attributes for achieving higher V_oc and potentially fine-tuning active-layer morphology. Herein, the polymer donor PBTI-FR was incorporated as the third component into the widely used PM6:L8-BO system,^30,47 constructing donor 1:donor 2:acceptor (D1:D2:A)-type TOSCs. It was found that PBTI-FR exhibits excellent compatibility with PM6, located within the PM6 region of the active layer to form a stable alloy-type donor structure. Compared to PM6, PBTI-FR features a deeper HOMO energy level and lower recombination energy, effectively reducing non-radiative recombination losses and thus enhancing V_oc. Furthermore, the addition of PBTI-FR significantly enhances the crystallinity of the ternary film while reducing defect density in the active layer. These improvements facilitate more efficient exciton dissociation, charge transport, and extraction. Ultimately, the TOSCs achieved a remarkable PCE of 20.52%, marking one of the highest efficiencies reported for OSCs. Additionally, the ternary devices demonstrated superior thermal stability compared to binary systems, retaining over 92% of their initial performance after 1008 h of heating. This study demonstrates a powerful strategy for next-generation organic photovoltaics by leveraging alloy polymer donors with lower energy loss and fine-tuned film morphology to achieve both high efficiency and exceptional stability.

Results and discussion

The chemical structures of PBTI-FR, PM6, and L8-BO are depicted in Fig. 1a, with the synthetic route to PBTI-FR detailed in the ESI† (Fig. S1–S8). The number-average molecular weights (M_ns) of PBTI-FR and PM6 were determined to be 65.2 and 62.0 kDa, respectively, by high-temperature gel permeation chromatography. The UV-vis absorption spectra of PBTI-FR, PM6, and L8-BO in both dilute chloroform solutions and film states are illustrated in Fig. 1b and Fig. S9 (ESI†). In blend films, the distinct absorption features of the donor and acceptor remain distinguishable, with maximum absorption peaks recorded at 613, 625, and 624 nm for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO systems, respectively (Fig. 1c). The incorporation of the third component, PBTI-FR, into the ternary blend films extends the optical absorption range, which is beneficial for enhancing the short-circuit current density (J_sc) of the corresponding devices. We selected a 560 nm excitation wavelength for the photoluminescence (PL) characterization, at which the PL emission from L8-BO excitation is nearly negligible (Fig. S10, ESI†), effectively excluding the interference of hole transfer from acceptor to donor. The PL spectra (Fig. S11, ESI†) revealed PL quenching efficiencies of 98.7%, 99.1%, and 99.2% for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO systems, respectively. These high quenching efficiencies indicate efficient exciton dissociation into charge carriers, contributing to improved PCEs in OSCs. The HOMO/LUMO (lowest unoccupied molecular orbital) energy levels of PBTI-FR, PM6 and L8-BO are characterized by cyclic voltammetry (CV) (Fig. S12, ESI†), found to be −5.57/−2.82, −5.49/−3.01, and −5.67/−3.97 eV, respectively (Fig. 1d). These well-aligned energy levels can facilitate efficient charge separation and transport in the ternary system. We further conducted TGA and DSC tests on both PM6 and PBTI-FR, with the results presented in Fig. S13 (ESI†). The TGA measurements unveil that the thermal decomposition temperatures corresponding to 5% weight loss for PM6 and PBTI-FR are 436 and 440 °C, respectively, underscoring the exceptional thermal stability of both materials.

Fig. 1 (a) Molecular structures of PM6, PBTI-FR, and L8-BO. (b) Normalized UV-vis absorption spectra of neat PM6, PBTI-FR, and L8-BO films, and (c) corresponding spectra for their blend films. (d) Energy level diagram of PM6, PBTI-FR, and L8-BO. (e) Contact angle images of neat PM6 and PBTI-FR films. (f) Interfacial energy and miscibility coefficients for PBTI-FR:PM6, PBTI-FR:L8-BO and PM6:L8-BO. (g) HOMO energy levels of PBTI-FR, PM6:PBTI-FR and PM6 determined by UPS measurement.

The precise spatial distribution of the third component within ternary blend films plays a pivotal role in determining the device's working mechanism.⁴⁵ Contact angle measurements were then conducted to evaluate material wettability and miscibility (see the ESI† for a detailed calculation process). Water and glycol droplet contact angles on the film surfaces (Fig. 1e and Fig. S14, ESI†) were used to calculate surface energy values, which were 25.70, 26.43, 26.12, and 27.24 mJ m⁻² for PBTI-FR, PM6, PBTI-FR:PM6 and L8-BO, respectively. The interfacial energy (γ_X–Y, Fig. 1f) between PBTI-FR and PM6 was calculated to be 0.008 mJ m⁻², significantly lower than the values for PBTI-FR:L8-BO (0.037 mJ m⁻²) and PM6:L8-BO (1.010 mJ m⁻²). The wettability coefficient (ω_T) was derived using the Young's equation: ω_T = (γ_T-A-γ_T-D)/γ_D–A. The calculated ω_T of 2.79 (ω_T > 1) indicates that PBTI-FR resides within the PM6-rich domain,⁵³ forming an alloy donor in the ternary blend.

Material miscibility strongly influences the nanomorphology of the active-layer films.²⁰ The Flory–Huggins interaction parameter (χ) serves as a measure of the interaction between components, with the χ values calculated as 0.0051, 0.0063, and 0.0226 for PM6:PBTI-FR, PM6:L8-BO, and PBTI-FR:L8-BO, respectively (Fig. 1f). A lower χ value indicates a better miscibility, yielding increased donor:acceptor interfaces, which are beneficial for exciton dissociation. In contrast, higher χ values may lead to excessive phase purity, which is detrimental to charge separation.^54,55 The exceptionally low χ value for PM6:PBTI-FR is particularly promising for forming an alloy donor phase in the ternary system. All the data above are summarized in Tables S1 and S2 (ESI†). To further confirm alloy donor generation, ultraviolet photoelectron spectroscopy (UPS) was performed to examine the HOMO levels of the donors (Fig. S15, ESI†). As shown in Fig. 1g, the HOMO levels of PBTI-FR, the PBTI-FR : PM6 (0.1 : 0.9) blend and PM6 were −5.22, −5.18 and −5.14 eV, respectively. The intermediate HOMO level of the PBTI-FR:PM6 blend further corroborates the formation of the alloy donor phase in the ternary system.

Theoretical calculations were conducted to elucidate the impact of incorporating PBTI-FR as the third component on intermolecular interactions within the ternary blend system. Density functional theory (DFT) calculations at the Y3LYP-D3/6-31G level were employed to determine the optimized molecular topologies of dimeric configurations for polymers PBTI-FR and PM6, as well as the acceptor L8-BO (Fig. S16, ESI†). The electrostatic potential (ESP) distributions along the molecular backbones of PM6 and PBTI-FR exhibited complementarity to those of L8-BO (Fig. S17, ESI†), enabling robust dipole–dipole interactions between neighboring donor–acceptor pairs.⁵⁶ Furthermore, DFT-calculated dipole moments for PBTI-FR, PM6 and L8-BO were 4.94, 0.88 and 2.99 Debye (D), respectively (Fig. 2a–c), with PBTI-FR exhibiting the highest dipole moment, indicating its propensity to form stronger intermolecular interactions.⁵⁷ To further analyze the stacking characteristics, four potential π–π dimeric configurations were simulated, and their Gibbs free energies , binding energies and interaction energies were compared (Fig. S18 and Table S3, ESI†). The average for the PM6:PM6, PM6:PBTI-FR and PBTI-FR:PBTI-FR complexes are −35.46, −35.00 and −35.95 kcal mol⁻¹, respectively. These values are relatively close, suggesting consistently favorable intermolecular interactions among the different molecular combinations. Importantly, such low values underscore the enhanced affinity and robust molecular packing between donor molecules, thereby promoting optimal and stable phase separation within the active layers.⁵⁸ The simulations revealed that the average binding and interaction energy values between PBTI-FR and PM6 exceed those between PM6 and PM6 in complex models (Fig. 2d). This enhanced interaction facilitated more effective molecular stacking, contributing to improved crystallinity and optimized film morphology, thereby favoring the formation of efficient hole transport pathways.⁵⁹

Fig. 2 DFT-optimized molecular conformations and corresponding dipole moments of (a) PBTI-FR, (b) PM6, and (c) L8-BO. (d) Average interaction and binding energies for PM6:PM6 and PM6:PBTI-FR complexes. (e) MD snapshots illustrating the initial and equilibrium states of the PM6:PBTI-FR:L8-BO ternary system. (f) Calculated end-to-end distances for binary and ternary systems. Calculated average (g) Coulombic and (h) van der Waals forces between L8-BO molecules within binary and ternary systems.

Molecular dynamics (MD) simulations were employed to investigate the intermolecular interactions within the active layers.⁶⁰ MD steady-state snapshots of initial and equilibrium states unveiled molecular miscibility and aggregation behavior in the PBTI-FR:L8-BO, PM6:L8-BO and PM6:PBTI-FR:L8-BO (Fig. 2e and Fig. S19, ESI†). End-to-end distances of molecules in each system were also determined by MD (Fig. 2f). In the ternary system, the end-to-end distances of PBTI-FR and PM6 chains were found to be 4.01 and 3.51 nm, respectively, which are shorter than those observed in the PBTI-FR:L8-BO (PBTI-FR: 4.30 nm) and PM6:L8-BO (PM6: 3.54 nm) binary systems. This reduction in end-to-end distance indicates enhanced aggregation of PM6 molecules induced by PBTI-FR, facilitating the formation of well-structured nanofibers and improving π–π stacking interactions.²¹ Furthermore, the acceptor L8-BO exhibited a minimum end-to-end distance of 1.66 nm in the PBTI-FR:L8-BO system. Statistical analysis of intermolecular interactions (Fig. 2g and h) revealed that the average Coulombic and van der Waals forces between L8-BO molecules were significantly higher in the PBTI-FR:L8-BO (43 356/−29 902 kJ mol⁻¹) and PM6:PBTI-FR:L8-BO (36 925/−24 902 kJ mol⁻¹) systems compared to the PM6:L8-BO (36 822/−24 682 kJ mol⁻¹) one. These elevated interaction forces correspond to reduced intermolecular distances and enhanced molecular packing of L8-BO within the ternary system. Notably, the stronger intermolecular interaction energy of L8-BO within the PBTI-FR:L8-BO blend likely promotes the formation of larger pure acceptor domains in the film.⁶¹

Based on a thorough understanding of the molecular interactions, we next investigated the impact of incorporating PBTI-FR into both the active-layer blends. The detailed fabrication processes for these devices are provided in the ESI.†Fig. 3a illustrates the current density–voltage (J–V) characteristics with associated photovoltaic parameters summarized in Table 1. For the optimal binary devices, the PM6:L8-BO system achieved a PCE of 18.75%, while PBTI-FR:L8-BO reached 16.98%. Notably, the PBTI-FR:L8-BO-based device delivered a higher V_oc of 0.91 V compared to that (0.86 V) of the PM6:L8-BO-based cell. This enhancement in V_oc was attributed to the deeper HOMO level of PBTI-FR relative to PM6.

Fig. 3 (a) J–V characteristics of the binary and ternary OSCs under AM 1.5 G at 100 mW cm⁻². (b) EQE spectra and the integral J_sc curves for the corresponding binary and ternary devices. (c) Statistical plots of PCE and FF of binary and ternary OSCs (15 devices). (d) FF values as a function of V_oc for this work and previously reported devices. (e) A comparative overview of the PCE value in this work against prior reports. The FTPS-EQE and EL spectra of the (f) PBTI-FR:L8-BO and (g) PM6:PBTI-FR:L8-BO-based devices, respectively. (h) EQE_ELs for the binary and ternary OSCs. (i) Diagram of the relevant electronic transitions between the ground state (S₀) and the lowest singlet excited state (S₁), serving to determine the recombination energy. (j) Schematic illustration of the operating principles and charge transfer dynamics for alloy donors. (k) Summary of the energy loss for the binary and ternary OSCs.

Table 1 Photovoltaic performance parameters of the PBTI-FR:L8-BO, PM6:L8-BO and PM6:PBTI-FR:L8-BO-based devices

Active layer^a	V oc (V)	J sc (mA cm^−2)	Cal. Jsc (mA cm^−2)^b	FF (%)	PCE (%)^c
a The device area is 6.4 mm^2. b Integrated current density obtained from EQE spectra. c Average values with standard deviation from 15 devices. d Certified results from the China National Center of Metrology.
PBTI-FR:L8-BO	0.91	24.07	23.66	77.77	16.98 (16.50 ± 0.33)
PM6:L8-BO	0.86	26.81	25.71	80.97	18.75 (18.34 ± 0.32)
PM6:PBTI-FR:L8-BO	0.89	28.07	26.91	82.55	20.52 (20.30 ± 0.10)
PM6:PBTI-FR:L8-BO^d	0.89	28.12	—	79.10	19.92

---

Inspired by these findings, we introduced PBTI-FR as a third component into the PM6:L8-BO-based system, constructing a ternary device based on the D1:D2:A type (the optimal ratio of PM6 : PBTI-FR was 0.9 : 0.1, Table S4, ESI†) active layer. Encouragingly, the resulting ternary device achieved an impressive PCE of 20.52% (certified PCE, 19.92%, Fig. S20, ESI†), ranked among the highest ones reported for OSCs (Table S5, ESI†). As anticipated, the V_oc rose from 0.86 V in the PM6:L8-BO binary device to 0.89 V in the PM6:PBTI-FR:L8-BO ternary device, reflecting a potential reduction in energy loss. Additionally, the J_sc of the ternary cell was increased to 28.07 mA cm⁻². Notably, the FF was elevated from 80.97% (PM6:L8-BO) to 82.55% (PM6:PBTI-FR:L8-BO), marking one of the highest FF values reported thus far. The external quantum efficiency (EQE) spectra and the integral J_sc curves (Fig. 3b) reveal the photocurrent derived from integrating the EQE curve, which exhibits a mismatch of less than 5% relative to the values obtained from the J–V measurements. Fig. 3c displays statistical box plots for PCE and FF, highlighting the excellent reproducibility of the ternary devices. Moreover, the theoretical maximum boundary of the FF affected by V_oc was established (Fig. 3d), followed by a statistical analysis of FF values reported in the literature. Notably, the FF of the PM6:PBTI-FR:L8-BO-based ternary device was close to the empirically predicted maximum FF (FF_max), indicating minimal losses through charge transport or recombination. Meanwhile, Fig. 3e presents reported PCE values for ternary cells, providing a comparative perspective on current advances in organic photovoltaics.

To systematically elucidate the impact of active layer thickness on the PCE of the ternary OSCs, we have accordingly fabricated devices with varying film thicknesses.⁶² As detailed in Fig. S21 and Table S6 (ESI†), ternary devices with active layer thicknesses of 253, 328 and 409 nm achieved PCEs of 18.80%, 18.14% and 17.55%, respectively. These results indicate that the device performance remains remarkably robust over a wide thickness range, thereby validating the scalability of the ternary strategy.

To evaluate the impact of incorporating the PBTI-FR component on energy loss (E_loss), we examined the optical band gaps (E_gs) of the PM6:L8-BO, PBTI-FR:L8-BO, and PM6:PBTI-FR:L8-BO blend films (1.430, 1.432, and 1.432 eV, respectively) by analyzing their electroluminescence (EL) and EQE spectra (Fig. S22, ESI†). According to the SQ limit, E_loss includes three terms: ΔE₁, ΔE₂, and ΔE₃. Here, ΔE₁ (0.262–0.263 eV) stems directly from the fundamental bandgap and was intrinsically unavoidable. ΔE₂ arises from radiative recombination below the bandgap and likewise cannot be eliminated. The radiative V_oc limit was calculated by combining the Fourier-transform photocurrent spectroscopic extra-quantum efficiency (FTPS-EQE) and EL spectra (Fig. S23 and Fig. 3f, g, ESI†), indicating the very small ΔE₂ below bandgaps (Table 2). Minor variations in ΔE₂ were observed across devices, with PBTI-FR:L8-BO exhibiting the smallest value of 0.058 eV. From the FTPS-EQE fitting, the obtained Urbach energies (E_us) for PM6:L8-BO, PBTI-FR:L8-BO, and PM6:PBTI-FR:L8-BO (19.74, 18.83, and 19.02 meV, respectively) reveal diminished energetic disorder upon PBTI-FR incorporation, thus lowering energy losses and improving charge-carrier mobility.⁶³ The third term, ΔE₃, signifies non-radiative recombination losses. ΔE₃ can be calculated using the formula ΔE₃ = −k_BT ln(EQE_EL), and the electroluminescence quantum efficiency (EQE_EL) is shown in Fig. 3h. The calculation revealed that the PBTI-FR:L8-BO system achieved the lowest ΔE₃ value of 0.212 eV. Moreover, incorporating PBTI-FR as a third component into the PM6:L8-BO system reduced ΔE₃ from 0.242 eV to 0.227 eV. Non-radiative recombination was related to electron-vibration coupling, and the recombination energy (λ) for the donor materials PBTI-FR and PM6 were determined from the ground to excited states (Fig. 3i), yielding values of 0.3088 eV and 0.3164 eV, respectively. The lower recombination energy of PBTI-FR enhanced the hybridization between the charge transfer (CT) state and the highly luminescent localized excited state, thereby mitigating non-radiative decay (Fig. 3j).⁶⁴ Consequently, PBTI-FR exhibited a lower non-radiative recombination energy loss, resulting in lower ΔE₃ in the ternary device. A detailed comparison of E_loss, summarized in Fig. 3k and Table 2, shows that PM6:L8-BO, PM6:PBTI-FR:L8-BO, and PBTI-FR:L8-BO devices exhibit E_loss values of 0.565, 0.549, and 0.533 eV, respectively. Therefore, incorporating PBTI-FR into the ternary system effectively lowers energy losses and results in an increased V_oc.

Table 2 Energy loss of the ternary and binary devices and the recombination energy for donors

Active layer	E g (eV)	ΔE1 (eV)	ΔE2 (eV)	ΔE3 (eV)	E loss (eV)	Donor	E I (eV)	E II (eV)	λ(Ereo) (eV)
PM6:L8-BO	1.430	0.262	0.061	0.242	0.565	PBTI-FR	0.1714	0.1375	0.3088
PM6:PBTI-FR:L8-BO	1.432	0.262	0.060	0.227	0.549	—	—	—	—
PBTI-FR:L8-BO	1.432	0.263	0.058	0.212	0.533	PM6	0.1710	0.1454	0.3164

---

We conducted an in-depth morphological analysis to elucidate the photovoltaic performance variations among the devices. Two-dimensional grazing incidence wide-angle X-ray scattering (2D-GIWAXS) was employed to assess the influence of PBTI-FR on intermolecular stacking in the ternary films. Fig. S24 (ESI†) presents the 2D-GIWAXS patterns of neat films, with the associated data summarized in Table S7 (ESI†). As shown in Fig. 4a, a distinct lamellar stacking peak was observed in the in-plane (IP) direction for the neat films, with the acceptor peak located at q_xy ≈ 0.43–0.44 Å⁻¹. In the out-of-plane (OOP) direction, a pronounced π–π stacking peak indicated the formation of vertical charge transport channels,⁶⁵ suggesting a preferential face-on orientation for both the donor and acceptor in pure membranes. Compared to PM6, PBTI-FR exhibited a larger crystal coherence length (CCLs) in the IP direction, signifying enhanced crystallinity and reduced disorder.

Fig. 4 (a) In-plane and out-of-plane 2D-GIWAXS line-cut profiles of neat films. (b)–(d) 2D-GIWAXS patterns. (e) In-plane and out-of-plane 2D-GIWAXS line-cut profiles of blend films. (f)–(h) AFM height images and (i)–(k) TEM micrographs of the binary and ternary blend films.

Fig. 4b–d demonstrate that all blend films retained a preferred face-on orientation. In the PBTI-FR:L8-BO blend, the PBTI-FR-based stacking peak appeared at 0.27 Å⁻¹ in the IP direction, while PM6:L8-BO showed a PM6-based peak at 0.30 Å⁻¹. Upon the addition of PBTI-FR to the PM6:L8-BO film, the IP (100) stacking peak shifted to 0.29 Å⁻¹, indicating the emergence of a hybrid crystalline phase.⁴⁷ Moreover, the PBTI-FR:L8-BO blend exhibited a more pronounced IP peak at 0.44 Å⁻¹ (Fig. 4e), aligning with the peak position of the acceptor and signifying robust molecular stacking. Intensity enhancements of both lamellar and π–π stacking peaks in the ternary film underscore the critical role of PBTI-FR in refining molecular orientation. The CCLs for the PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO blends were calculated as 3.6, 2.5, and 3.4 nm in the OOP direction, and 13.5, 7.9, and 10.7 nm in the IP direction, respectively. These results correlate well with the observed trend in E_u, highlighting that PBTI-FR incorporation enhances film crystallinity and molecular stacking, thereby improving the J_sc and FF.

To reconcile these observations, the morphological features were further investigated via atomic force microscopy (AFM) and transmission electron microscopy (TEM). As shown in Fig. 4f–h and Fig. S25 (ESI†), the surface roughness of the active layers for the PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO systems was 1.35, 1.06, and 0.99 nm, respectively. The decreased roughness in the ternary blend indicated a more homogeneous surface, which likely minimizes the interfacial defects.⁶⁶ Intuitively, the ternary system formed a distinct nanofiber structure, suggesting that the addition of PBTI-FR facilitated fine-tuning of the film morphology and phase separation, thereby improving device efficiency.⁴⁶ In contrast, the PBTI-FR:L8-BO blend film appeared to have more surface defects and larger pure-phase domains, presumably due to excessive aggregation from strong intermolecular interactions.^59,67 TEM images (Fig. 4i–k) corroborate the AFM results. The bright donor-enriched and darker acceptor-enriched regions reveal the non-uniform distribution and pronounced molecular aggregation in the PBTI-FR:L8-BO film.⁶⁸ Meanwhile, the ternary blend exhibited the most uniform donor–acceptor distribution. Together with theoretical calculations, these observations suggested that adding PBTI-FR into the PM6:L8-BO system could optimize donor–acceptor ordering, forming nanofiber networks that facilitate exciton dissociation and charge transport and thus improving the J_sc and FF.

The trap density of states (tDOS) for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based devices was characterized using thermal admittance spectroscopy (TAS),⁶⁹ as presented in Fig. 5a. For the binary blends, PBTI-FR:L8-BO demonstrated a lower tDOS below an energy depth of 0.43 eV, while PM6:L8-BO showed a reduced tDOS above this threshold. The incorporation of PBTI-FR in the PM6:PBTI-FR:L8-BO system synergistically diminished the tDOS across all energy depths, achieving the lowest overall defect density compared to the binary counterparts. This reduction in trap states significantly contributed to improving the FF.⁷⁰ To gain deeper insights into the differences in J_sc and FF between the ternary and binary systems, we performed in detail the exciton dynamics process, and the data are summarized in Fig. 5b and Table S8 (ESI†). The photocurrent density (J_ph) and effective voltage (V_eff) measurements of the optimal devices (Fig. S26, ESI†) revealed that the J_ph approached saturation at V_eff > 1 eV, indicating efficient exciton dissociation into free charge carriers. The exciton dissociation efficiency (P_diss) and charge collection efficiency (P_coll) for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based devices were evaluated to be 94.72%/80.07%, 95.85%/88.25%, and 98.02%/90.32%, respectively. These high efficiencies in the ternary device underscore enhanced exciton dissociation and charge extraction capabilities, attributable to its reduced defect density.⁷¹

Fig. 5 (a) Trap density of states spectra of binary and ternary devices. (b) Summary of carrier dynamics parameters of the binary and ternary devices. (c) TPC and (d) TPV curves of the binary and ternary devices. (e) Normalized TRPL dynamics of binary and ternary blend films. (f) Operation stability under MPP tracking at 100 mW cm⁻² with white LED. (g) Normalized device stability evolution of the binary and ternary devices in a nitrogen atmosphere under 65 °C heating.

To examine the charge recombination in binary and ternary devices, the dependence of J_sc and V_oc on incident light power (P_light, Fig. S27, ESI†) was evaluated by the equation of (J_sc ∝ P_light^α) and (V_oc ∝ αkT/q ln(P_light)), in which k, T and q are the Boltzmann constant, absolute temperature, and elementary charge, respectively. The parameters α(J_sc) and α(V_oc) values for the corresponding binary and ternary devices were 0.956/1.26, 0.983/1.20 and 0.998/1.10, respectively. The α(J_sc) and α(V_oc) values of the ternary devices were closer to unity than those of binary devices, indicating weaker bimolecular recombination and less monomolecular recombination due to its low defect density inhibiting trap-assisted recombination.⁷² Charge carrier mobilities were examined using the space-charge-limited current (SCLC) method. To accurately assess the charge carrier mobilities, we determined the relative dielectric constants (ε_r) for binary and ternary blend films using impedance spectroscopy (Fig. S28, ESI†).⁷³ the PBTI-FR:L8-BO blend exhibits the highest relative dielectric constant (ε_r) of 4.32, whereas the PM6:L8-BO blend shows the lowest value at 3.86, with the ternary blend falling in between (3.96). This observation confirms that the incorporation of PBTI-FR into the PM6:L8-BO matrix enhances the dielectric properties of the active layer. This enhancement is beneficial to facilitating exciton dissociation and carrier extraction rate while reducing monomolecular recombination, ultimately leading to improved performance of PM6:PBTI-FR:L8-BO-based TOSCs.^74–76 The calculated hole mobility (μ_h) and electron mobility (μ_e) for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based devices were found to be 4.86 × 10⁻⁴/3.48 × 10⁻⁴, 2.97 × 10⁻⁴/3.26 × 10⁻⁴ and 3.25 × 10⁻⁴/3.29 × 10⁻⁴ cm² V⁻¹ s⁻¹ respectively (Fig. S29 and Table S9, ESI†), with μ_h/μ_e ratios of 1.40, 0.91 and 0.99. Although PBTI-FR:L8-BO exhibited superior mobilities due to pronounced phase separation and crystallinity, high μ_h/μ_e ratios could cause charge imbalance and hole accumulation, increasing the probability of bimolecular recombination, detrimentally affecting J_sc and FF.⁷⁷ The addition of PBTI-FR optimized the crystallinity and phase separation in the ternary system, resulting in improved hole and electron mobilities and more balanced charge transport than the PM6:L8-BO-based device, thereby boosting J_sc and FF.

Transient photocurrent (TPC) and transient photovoltage (TPV) measurements were conducted to evaluate charge recombination dynamics.²⁰ As shown in Fig. 5c, the ternary device demonstrated faster rise and decay processes. TPC fitting revealed charge extraction times of 0.19 μs (PBTI-FR:L8-BO), 0.17 μs (PM6:L8-BO), and 0.06 μs (PM6:PBTI-FR:L8-BO), indicating more efficient charge generation and extraction with fewer traps in the ternary system. The carrier lifetimes (τ) derived from fitting the TPV curves (Fig. 5d) were 1.20, 1.42 and 2.05 μs for the PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based systems, suggesting extended carrier lifetimes in the ternary system. Time-resolved photoluminescence (TRPL) confirmed these results (Fig. 5e and Fig. S30, ESI†). The neat films of PBTI-FR, PM6, and L8-BO exhibited average TRPL lifetimes of 1.05, 0.81, and 1.59 ns, respectively (Table S10, ESI†), whereas the blend films exhibited nonradiative lifetimes of 0.69, 0.80, and 1.07 ns (Table S11, ESI†). The extended lifetimes in the neat films imply more efficient exciton dissociation at the donor–acceptor interfaces.⁷⁸ In addition, the enhanced nonradiative lifetimes of blend films can reduce charge recombination, thereby improving J_sc and FF and enhancing long-term device stability.^79,80

To assess the device stability, maximum power point (MPP) tracking was carried out on the PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based cells (Fig. 5f). The results demonstrated stable output over an extended period, highlighting the robust performance of these systems.^81–83 Further thermal aging tests at 65 °C heating for periods exceeding 1000 h (Fig. 5g and Tables S12–S14, ESI†) revealed that the PM6:PBTI-FR:L8-BO ternary system retained over 92.2% of its initial efficiency, outperforming the respective 84.3% and 84.9% retentions observed in the PBTI-FR:L8-BO and PM6:L8-BO systems. A subsequent 500-hour storage stability examination showed that the devices based on PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO maintained 94.9%, 96.9%, and 97.2% of their original performance, respectively (Fig. S31, ESI†). These findings confirm that the ternary formulation of PM6:PBTI-FR:L8-BO significantly boosts both thermal and shelf stability, effectively mitigating performance degradation under prolonged operational and storage conditions. The enhanced stability can be attributed to the optimized morphology⁴⁸ and well-balanced electronic properties⁴² imparted by the incorporation of the third component PBTI-FR.

Transient absorption (TA) spectroscopy using a pump wavelength of 800 nm was conducted to elucidate the charge transfer processes in the binary and ternary devices. This pump wavelength was intentionally chosen to selectively excite the acceptor in the blend films, thus minimizing donor absorption interference. Fig. S32–S35 (ESI†) presented the 2D-TA and TA spectra at various delay times for donor-only and acceptor-only films, respectively. The ground state bleaching (GSB) signals for donors emerged at 500–650 nm, while the GSB for acceptors spanned 620–850 nm. In Fig. 6a–c, the blend films revealed a pronounced decay feature at ∼850 nm corresponding to the acceptor L8-BO GSB, along with a peak at 900 nm attributed to L8-BO excited state absorption (ESA). Notably, the intensity of the donor GSB at ∼595 nm was positively correlated with the amount of hole transfer from accepter L8-BO to the donor, thereby serving as a direct indicator of exciton dissociation efficiency within the hybrid film.⁴⁷ Among the films investigated, the ternary film displayed the strongest GSB signal at ∼595 nm, signifying superior exciton dissociation. Conversely, the PBTI-FR:L8-BO binary blend showed the weakest GSB intensity in this region, indicative of less efficient dissociation efficiency, aligning with the J_ph–V_eff results. As illustrated in Fig. 6d–f, the TA kinetic curves demonstrated that as the acceptor GSB around 820 nm decayed, the donor GSB near 580 nm subsequently rose in all blends, indicating that the process of photo-induced hole transfer from L8-BO to the donor.⁸⁴

Fig. 6 Femtosecond TA spectra of (a) PBTI-FR:L8-BO, (b) PM6:L8-BO and (c) PM6:PBTI-FR:L8-BO blend films. Decay dynamics probed at different wavelengths in the (d) PBTI-FR:L8-BO, (e) PM6:L8-BO and (f) PM6:PBTI-FR:L8-BO blend films.

TA spectra at various delay times (Fig. S36, ESI†) revealed kinetic parameters τ₁ and τ₂, representing the rapid exciton dissociation at the heterojunction interface and exciton diffusion into the interface within crystalline domains.⁸⁵τ₁ and τ₂ were extracted by fitting a double exponential function to the kinetic signals around 580 nm using Surface Xplorer software. The resulting τ₁/τ₂ values for PBTI-FR:L8-BO, PM6:L8-BO, and PM6:PBTI-FR:L8-BO-based devices were determined to be 1.98/257, 0.59/277 and 0.44/183 ps, respectively. The notably reduced τ₁ in the ternary system suggested that the incorporation of PBTI-FR strengthens donor–acceptor interactions, thus facilitating rapid exciton dissociation.⁸⁶ Meanwhile, the decrease in τ₂ indicated increased diffusion constants, which were associated with improved crystalline stacking and refined phase separation in the ternary blend. These results confirmed that PBTI-FR addition promoted exciton dissociation and diffusion, ultimately contributing to the improved J_sc and FF observed in the ternary device.

To evaluate the generality of the PBTI-FR donor, we fabricated a series of ternary devices employing alternative polymer donor/acceptor systems (PM6:BTP-ec9, PM6:Y6 and D18-Cl:Y6), with their photovoltaic performance parameters detailed in Fig. S37 and Table S15 (ESI†). The incorporation of the third component PBTI-FR consistently enhanced both the J_scs and FFs of OSCs, leading to higher PCEs compared to binary control devices. Notably, the PM6:BTP-ec9 system processed with PBTI-FR achieved an outstanding PCE of 19.96%, surpassing the benchmark efficiency of 18.57%. These results not only underscore the excellent applicability of PBTI-FR but also highlight the pivotal role of the dual-donor ternary strategy in achieving high-performance OSCs.

Conclusion

In this work, a new BTI-based polymer donor (PBTI-FR) is designed and incorporated into the PM6:L8-BO blend, and the resulting TOSCs show a dual-polymer-donor phase. By leveraging strong dipole–dipole interactions, PBTI-FR can effectively tune the active-layer morphology and energy levels, facilitating the formation of alloy donors. This approach deepens the HOMO level, reduces trap-assisted recombination, and enhances crystallinity in the ternary blend, promoting efficient exciton dissociation and balanced charge transport. As a result, the TOSC delivers a remarkable PCE of 20.52%, among the highest ones reported for organic photovoltaics, with an exceptional fill factor exceeding 82%. In-depth and comprehensive material and device characterizations reveal the accelerated exciton dissociation kinetics and increased diffusion constant in the ternary system, underlining the positive impact of PBTI-FR on charge generation and collection. Moreover, the ternary device exhibits a superior thermal stability, retaining over 92% of its initial efficiency after 1008 h under continuous heating. This work provides important insights into how the delicate design and incorporation of a polymer donor with tailored energy levels and strong intermolecular interactions can increase both device efficiency and stability, offering a powerful approach for the next-generation organic photovoltaics.

Data availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

X. G. acknowledges the financial support from the Shenzhen Innovation Commission of Science and Technology (KCXST20221021111413031). B. L. is thankful for the financial support of the Guangdong Basic and Applied Basic Research Foundation (2023A1515011048). This work was supported by the National Research Foundation of Korea (RS-2023-00217884 and RS-2024-00432362). The authors are also grateful for the technical support from SUSTech Core Research Facilities and the Computational Science and Engineering of Southern University of Science and Technology.

References

1. H. J. Hovel, Sol. Energy Mater., 1980, 2, 277–312.
2. G. Li, R. Zhu and Y. Yang, Nat. Photonics, 2012, 6, 153–161.
3. J. Hou, O. Inganäs, R. H. Friend and F. Gao, Nat. Mater., 2018, 17, 119–128.
4. E. Trippodo, V. Campisciano, L.-W. Feng, Y. Chen, W. Huang, J. M. Alzola, D. Zheng, V. K. Sangwan, M. C. Hersam, M. R. Wasielewski, B. Pignataro, F. Giacalone, T. J. Marks and A. Facchetti, J. Mater. Chem. C, 2023, 11, 8074–8083.
5. D. Qian, L. Ye, M. Zhang, Y. Liang, L. Li, Y. Huang, X. Guo, S. Zhang, Z. A. Tan and J. Hou, Macromolecules, 2012, 45, 9611–9617.
6. C. Wang, X. Ma, Y.-F. Shen, D. Deng, H. Zhang, T. Wang, J. Zhang, J. Li, R. Wang, L. Zhang, Q. Cheng, Z. Zhang, H. Zhou, C. Tian and Z. Wei, Joule, 2023, 7, 2386–2401.
7. Y. Liang, Z. Xu, J. Xia, S. T. Tsai, Y. Wu, G. Li, C. Ray and L. Yu, Adv. Mater., 2010, 22, E135–E138.
8. V. Vohra, K. Kawashima, T. Kakara, T. Koganezawa, I. Osaka, K. Takimiya and H. Murata, Nat. Photonics, 2015, 9, 403–408.
9. X. Guo, N. Zhou, S. J. Lou, J. Smith, D. B. Tice, J. W. Hennek, R. P. Ortiz, J. T. L. Navarrete, S. Li, J. Strzalka, L. X. Chen, R. P. H. Chang, A. Facchetti and T. J. Marks, Nat. Photonics, 2013, 7, 825–833.
10. Y. Liang, D. Zhang, Z. Wu, T. Jia, L. Lüer, H. Tang, L. Hong, J. Zhang, K. Zhang, C. J. Brabec, N. Li and F. Huang, Nat. Energy, 2022, 7, 1180–1190.
11. D. Baran, R. S. Ashraf, D. A. Hanifi, M. Abdelsamie, N. Gasparini, J. A. Röhr, S. Holliday, A. Wadsworth, S. Lockett, M. Neophytou, C. J. M. Emmott, J. Nelson, C. J. Brabec, A. Amassian, A. Salleo, T. Kirchartz, J. R. Durrant and I. McCulloch, Nat. Mater., 2016, 16, 363–369.
12. Y. Lin, J. Wang, Z. G. Zhang, H. Bai, Y. Li, D. Zhu and X. Zhan, Adv. Mater., 2015, 27, 1170–1174.
13. S. Xu, W. Wang, H. Liu, X. Yu, F. Qin, H. Luo, Y. Zhou and Z. A. Li, Macromol. Rapid Commun., 2022, 43, 2200276.
14. Z. Zheng, J. Wang, P. Bi, J. Ren, Y. Wang, Y. Yang, X. Liu, S. Zhang and J. Hou, Joule, 2022, 6, 171–184.
15. J. Ge, L. Hong, W. Song, L. Xie, J. Zhang, Z. Chen, K. Yu, R. Peng, X. Zhang and Z. Ge, Adv. Energy Mater., 2021, 11, 2100800.
16. Y. Wang, K. Sun, C. Li, C. Zhao, C. Gao, L. Zhu, Q. Bai, C. Xie, P. You, J. Lv, X. Sun, H. Hu, Z. Wang, H. Hu, Z. Tang, B. He, M. Qiu, S. Li and G. Zhang, Adv. Mater., 2024, 36, 2411957.
17. Z. Ren, S. Luo, X. Shi, Y. Dou, T. Liu, L. Wang, K. K. Tsang, F. Wang, Y. Zhao, Y. Liu, X. Hu, X. Peng, W. Liu, H. Yan and S. Chen, Sci. China: Chem., 2024, 67, 1941–1945.
18. L. Zhong, Z. Sun, S. Lee, S. Jeong, S. Jung, Y. Cho, J. Park, J. Park, S. J. Yoon and C. Yang, Adv. Funct. Mater., 2023, 33, 2305450.
19. 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.
20. 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.
21. C. Chen, L. Wang, W. Xia, K. Qiu, C. Guo, Z. Gan, J. Zhou, Y. Sun, D. Liu, W. Li and T. Wang, Nat. Commun., 2024, 15, 6865.
22. C. Li, G. Yao, X. Gu, J. Lv, Y. Hou, Q. Lin, N. Yu, M. S. Abbasi, X. Zhang, J. Zhang, Z. Tang, Q. Peng, C. Zhang, Y. Cai and H. Huang, Nat. Commun., 2024, 15, 8872.
23. H. Lu, D. Li, W. Liu, G. Ran, H. Wu, N. Wei, Z. Tang, Y. Liu, W. Zhang and Z. Bo, Angew. Chem., Int. Ed., 2024, 63, e202407007.
24. S. M. Menke, N. A. Ran, G. C. Bazan and R. H. Friend, Joule, 2018, 2, 25–35.
25. X. Zhang, C. Li, J. Xu, R. Wang, J. Song, H. Zhang, Y. Li, Y.-N. Jing, S. Li, G. Wu, J. Zhou, X. Li, Y. Zhang, X. Li, J. Zhang, C. Zhang, H. Zhou, Y. Sun and Y. Zhang, Joule, 2022, 6, 444–457.
26. Z. Chen, Y. Xiao, H. Yao, J. Ren, T. Zhang, J. Qiao, S. Zhu, R. Lin, X. Hao and J. Hou, Adv. Mater., 2024, 36, 2408858.
27. M. Koppe, H. J. Egelhaaf, G. Dennler, M. C. Scharber, C. J. Brabec, P. Schilinsky and C. N. Hoth, Adv. Funct. Mater., 2010, 20, 338–346.
28. H. Huang, L. Yang and B. Sharma, J. Mater. Chem. A, 2017, 5, 11501–11517.
29. Z. Fei, F. D. Eisner, X. Jiao, M. Azzouzi, J. A. Röhr, Y. Han, M. Shahid, A. S. R. Chesman, C. D. Easton, C. R. McNeill, T. D. Anthopoulos, J. Nelson and M. Heeney, Adv. Mater., 2018, 30, 1705209.
30. M. Zhang, L. Zhu, T. Hao, G. Zhou, C. Qiu, Z. Zhao, N. Hartmann, B. Xiao, Y. Zou, W. Feng, H. Zhu, M. Zhang, Y. Zhang, Y. Li, T. P. Russell and F. Liu, Adv. Mater., 2021, 33, 2007177.
31. F. Zhang, T. Dai, X. Li, M. Li, Y. Liu, D. Wang, D. Xu and R. Hu, Adv. Funct. Mater., 2024, 35, 2414260.
32. Z. Chen, W. Song, K. Yu, J. Ge, J. Zhang, L. Xie, R. Peng and Z. Ge, Joule, 2021, 5, 2395–2407.
33. H. Yao, Y. Cui, D. Qian, C. S. Ponseca, A. Honarfar, Y. Xu, J. Xin, Z. Chen, L. Hong, B. Gao, R. Yu, Y. Zu, W. Ma, P. Chabera, T. Pullerits, A. Yartsev, F. Gao and J. Hou, J. Am. Chem. Soc., 2019, 141, 7743–7750.
34. J. Zhang, H. S. Tan, X. Guo, A. Facchetti and H. Yan, Nat. Energy, 2018, 3, 720–731.
35. X. Liu, Y. Yan, Y. Yao and Z. Liang, Adv. Funct. Mater., 2018, 28, 1802004.
36. X. Duan, W. Song, J. Qiao, X. Li, Y. Cai, H. Wu, J. Zhang, X. Hao, Z. Tang, Z. Ge, F. Huang and Y. Sun, Energy Environ. Sci., 2022, 15, 1563–1572.
37. C. Zhu, S. Chung, J. Zhao, Y. Sun, B. Zhao, Z. Zhao, S. Kim, K. Cho and Z. Kan, Adv. Sci., 2023, 10, 2303150.
38. T. Huang, Y. Zhang, J. Wang, Z. Cao, S. Geng, H. Guan, D. Wang, Z. Zhang, Q. Liao and J. Zhang, Nano Energy, 2024, 121, 109226.
39. X. Li, A. Tang, H. Wang, Z. Wang, M. Du, Q. Guo, Q. Guo and E. Zhou, Angew. Chem., Int. Ed., 2023, 62, e202306847.
40. Z. Liu, M. Zhang, L. Zhang, S. Y. Jeong, S. Geng, H. Y. Woo, J. Zhang, F. Zhang and X. Ma, Chem. Eng. J., 2023, 471, 144711.
41. 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.
42. N. Gasparini, S. H. K. Paleti, J. Bertrandie, G. Cai, G. Zhang, A. Wadsworth, X. Lu, H.-L. Yip, I. McCulloch and D. Baran, ACS Energy Lett., 2020, 5, 1371–1379.
43. Y. Zhu, A. Gadisa, Z. Peng, M. Ghasemi, L. Ye, Z. Xu, S. Zhao and H. Ade, Adv. Energy Mater., 2019, 9, 1900376.
44. A. D. de Zerio and C. Müller, Adv. Energy Mater., 2018, 8, 1702741.
45. M. Günther, N. Kazerouni, D. Blätte, J. D. Perea, B. C. Thompson and T. Ameri, Nat. Rev. Mater., 2023, 8, 456–471.
46. L. Zhu, M. Zhang, J. Xu, C. Li, J. Yan, G. Zhou, W. Zhong, T. Hao, J. Song, X. Xue, Z. Zhou, R. Zeng, H. Zhu, C.-C. Chen, R. C. I. MacKenzie, Y. Zou, J. Nelson, Y. Zhang, Y. Sun and F. Liu, Nat. Mater., 2022, 21, 656–663.
47. J. Gao, N. Yu, Z. Chen, Y. Wei, C. Li, T. Liu, X. Gu, J. Zhang, Z. Wei, Z. Tang, X. Hao, F. Zhang, X. Zhang and H. Huang, Adv. Sci., 2022, 9, 2203606.
48. A. Liang, Y. Sun, S. Chung, J. Shin, K. Sun, C. Zhu, J. Zhao, Z. Zhao, Y. Zhong, G. Zhang, K. Cho and Z. Kan, Nano-Micro Lett., 2024, 17, 72.
49. B. Fan, W. Zhong, X. F. Jiang, Q. Yin, L. Ying, F. Huang and Y. Cao, Adv. Energy Mater., 2016, 7, 1602127.
50. S. Pang, W. Deng, L. Pan, X. Liu, Z. Shen, H. Li, P. Cheng, J. Zhu, W. Yan and C. Duan, iScience, 2025, 28, 111682.
51. Q. An, J. Wang, X. Ma, J. Gao, Z. Hu, B. Liu, H. Sun, X. Guo, X. Zhang and F. Zhang, Energy Environ. Sci., 2020, 13, 5039–5047.
52. J. Shi, K. Sun, Z. Chen, Y. Qiu, H. Liu, W. Ma, Q. Liu and Z. Ge, Angew. Chem., Int. Ed., 2024, 63, e202318360.
53. C. Wang, W. Zhang, X. Meng, J. Bergqvist, X. Liu, Z. Genene, X. Xu, A. Yartsev, O. Inganäs, W. Ma, E. Wang and M. Fahlman, Adv. Energy Mater., 2017, 7, 1700390.
54. Y. Bai, Z. Zhang, Q. Zhou, H. Geng, Q. Chen, S. Kim, R. Zhang, C. Zhang, B. Chang, S. Li, H. Fu, L. Xue, H. Wang, W. Li, W. Chen, M. Gao, L. Ye, Y. Zhou, Y. Ouyang, C. Zhang, F. Gao, C. Yang, Y. Li and Z.-G. Zhang, Nat. Commun., 2023, 14, 2926.
55. T. Gokulnath, K. Feng, H.-Y. Park, Y. Do, H. Park, R. D. Gayathri, S. S. Reddy, J. Kim, X. Guo, J. Yoon and S.-H. Jin, ACS Appl. Mater. Interfaces, 2022, 14, 11211–11221.
56. L. Wang, C. Chen, Y. Fu, C. Guo, D. Li, J. Cheng, W. Sun, Z. Gan, Y. Sun, B. Zhou, C. Liu, D. Liu, W. Li and T. Wang, Nat. Energy, 2024, 9, 208–218.
57. X. Ji, T. Wang, Q. Fu, D. Liu, Z. Wu, M. Zhang, H. Y. Woo and Y. Liu, Macromol. Rapid Commun., 2023, 44, 2300213.
58. X. He, C. C. S. Chan, X. Zou, S. Zhang, P. W. K. Fong, J. Kim, G. Li, X. Hu, W. Ma, K. S. Wong and W. C. H. Choy, Adv. Energy Mater., 2023, 13, 2203697.
59. C. Han, J. Wang, L. Chen, J. Chen, L. Zhou, P. Wang, W. Shen, N. Zheng, S. Wen, Y. Li and X. Bao, Adv. Funct. Mater., 2021, 31, 2107026.
60. B. Liu, L. Wenhao, L. Zhenmin, S. Sarvar, L. Fereidooni and A. Kasaeian, J. Mol. Liq., 2024, 414, 126052.
61. T. Dai, A. Tang, Z. He, M. Du, P. Lei, Q. Zeng, Z. Wang, Y. Wang, S. Lu, Y. Zhong and E. Zhou, Energy Environ. Sci., 2023, 16, 2199–2211.
62. G. Wang, M. A. Adil, J. Zhang and Z. Wei, Adv. Mater., 2018, 31, 1805089.
63. C. Kaiser, O. J. Sandberg, N. Zarrabi, W. Li, P. Meredith and A. Armin, Nat. Commun., 2021, 12, 3988.
64. Y. Shi, Y. Chang, K. Lu, Z. Chen, J. Zhang, Y. Yan, D. Qiu, Y. Liu, M. A. Adil, W. Ma, X. Hao, L. Zhu and Z. Wei, Nat. Commun., 2022, 13, 3256.
65. J. Rivnay, S. C. B. Mannsfeld, C. E. Miller, A. Salleo and M. F. Toney, Chem. Rev., 2012, 112, 5488–5519.
66. P. Yadav, H. Kim, T. Gokulnath, J. S. Yoo, M. J. Jeon, R. Kumaresan, H.-Y. Park and S.-H. Jin, ACS Appl. Mater. Interfaces, 2024, 16, 62082–62092.
67. T. Hao, W. Zhong, S. Leng, R. Zeng, M. Zhang, L. Zhu, Y. Yang, J. Song, J. Xu, G. Zhou, Y. Zou, Y. Zhang and F. Liu, Sci. China: Chem., 2022, 65, 1634–1641.
68. Q. An, F. Zhang, Q. Sun, M. Zhang, J. Zhang, W. Tang, X. Yin and Z. Deng, Nano Energy, 2016, 26, 180–191.
69. C. Melzer, E. J. Koop, V. D. Mihailetchi and P. W. M. Blom, Adv. Funct. Mater., 2004, 14, 865–870.
70. J. Zhou, D. He, Y. Li, F. Huang, J. Zhang, C. Zhang, Y. Yuan, Y. Lin, C. Wang and F. Zhao, Adv. Mater., 2022, 35, 2207336.
71. O. V. Mikhnenko, M. Kuik, J. Lin, N. van der Kaap, T. Q. Nguyen and P. W. M. Blom, Adv. Mater., 2013, 26, 1912–1917.
72. W. L. Leong, S. R. Cowan and A. J. Heeger, Adv. Energy Mater., 2011, 1, 517–522.
73. M. P. Hughes, K. D. Rosenthal, N. A. Ran, M. Seifrid, G. C. Bazan and T. Q. Nguyen, Adv. Funct. Mater., 2018, 28, 1801542.
74. L. J. A. Koster, S. E. Shaheen and J. C. Hummelen, Adv. Energy Mater., 2012, 2, 1246–1253.
75. N. Camaioni and R. Po, J. Phys. Chem. Lett., 2013, 4, 1821–1828.
76. J. Deng, B. Huang, W. Li, L. Zhang, S. Y. Jeong, S. Huang, S. Zhang, F. Wu, X. Xu, G. Zou, H. Y. Woo, Y. Chen and L. Chen, Angew. Chem., Int. Ed., 2022, 61, e202202177.
77. X. Liao, M. Liu, H. Pei, P. Zhu, X. Xia, Z. Chen, Y. Zhang, Z. Wu, Y. Cui, G. Xu, M. Gao, L. Ye, R. Ma, T. Liu, X. Lu, H. Zhu and Y. Chen, Angew. Chem., Int. Ed., 2024, 63, e202318595.
78. X. He, F. Qi, X. Zou, Y. Li, H. Liu, X. Lu, K. S. Wong, A. K. Y. Jen and W. C. H. Choy, Nat. Commun., 2024, 15, 2103.
79. H. Bi, X. Zuo, B. Liu, D. He, L. Bai, W. Wang, X. Li, Z. Xiao, K. Sun, Q. Song, Z. Zang and J. Chen, J. Mater. Chem. A, 2021, 9, 3940–3951.
80. G. Zhang, K. Zhang, Q. Yin, X.-F. Jiang, Z. Wang, J. Xin, W. Ma, H. Yan, F. Huang and Y. Cao, J. Am. Chem. Soc., 2017, 139, 2387–2395.
81. S. A. Gevorgyan, I. M. Heckler, E. Bundgaard, M. Corazza, M. Hösel, R. R. Søndergaard, G. A. dos Reis Benatto, M. Jørgensen and F. C. Krebs, J. Phys. D: Appl. Phys., 2017, 50, 103001.
82. M. Corazza, F. C. Krebs and S. A. Gevorgyan, Sol. Energy Mater. Sol. Cells, 2014, 130, 99–106.
83. S. A. Gevorgyan, M. V. Madsen, B. Roth, M. Corazza, M. Hösel, R. R. Søndergaard, M. Jørgensen and F. C. Krebs, Adv. Energy Mater., 2015, 6, 1501208.
84. M. Zhang, Z. Xiao, W. Gao, Q. Liu, K. Jin, W. Wang, Y. Mi, Q. An, X. Ma, X. Liu, C. Yang, L. Ding and F. Zhang, Adv. Energy Mater., 2018, 8, 1801968.
85. R. Zeng, L. Zhu, M. Zhang, W. Zhong, G. Zhou, J. Zhuang, T. Hao, Z. Zhou, L. Zhou, N. Hartmann, X. Xue, H. Jing, F. Han, Y. Bai, H. Wu, Z. Tang, Y. Zou, H. Zhu, C.-C. Chen, Y. Zhang and F. Liu, Nat. Commun., 2023, 14, 4148.
86. P. Ding, Z. Chen, D. Yang, X. Yu, J. Shi, Y. Chen, J. Zhu, J. Wu, X. Cao, L. Xie, F. Chen and Z. Ge, Adv. Mater., 2025, 37, 2414080.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ee00812c

‡ These authors contributed equally.

---