Organic solar cells (OSCs) are attracting significant attention due to their solution processability, cost-effectiveness, and mechanical flexibility. Recent advancements in non-fullerene acceptors (NFAs) have led to power conversion efficiencies (PCEs) exceeding 18%. However, these high-efficiency OSCs typically utilize active layers around 100 nm thick, hindering large-scale, cost-effective roll-to-roll production. Thicker active layers (several hundred nanometers) are desirable for improved processability, but usually lead to reduced efficiency due to challenges in exciton diffusion and charge recombination. This research addresses the critical challenge of creating efficient OSCs with thicker active layers. Exciton diffusion length is a key factor, as excitons must reach the donor-acceptor interface to dissociate into charge carriers. The typical exciton diffusion length (around 10 nm) is often mismatched with the phase separation scale in bulk heterojunction (BHJ) OSCs. Optimizing the vertical phase separation in thick active layers is crucial for enhancing the donor-acceptor interface area, facilitating charge transport, and suppressing charge recombination. This work aims to overcome these limitations and demonstrate high-performance thick-film OSCs.

The literature extensively documents the challenges of producing efficient thick-film organic solar cells. Many studies show a significant drop in power conversion efficiency as the active layer thickness increases beyond the optimal 100 nm. This is attributed to several factors, including insufficient exciton diffusion to reach the donor-acceptor interface, increased charge recombination, and imbalanced charge transport. Researchers have explored various strategies to overcome these limitations, including modifications to the active layer morphology, the use of ternary blends, and the incorporation of additives. Previous attempts at thick-film fullerene-based OSCs have yielded limited success, with the highest efficiency only reaching 11.3% at an active layer thickness of 280 nm. The limited success is largely due to the unbalanced hole and electron mobilities. This study builds on existing work by focusing on the optimization of both exciton diffusion and vertical phase separation through a combination of material selection and a novel layer-by-layer processing technique.

This study employed a ternary blend approach using the polymer donor PM6 and two non-fullerene acceptors, BTP-eC9 and L8-BO-F, selected for their complementary absorption and enhanced exciton diffusion length in the mixed phase. Conventional OSC architecture (ITO/PEDOT:PSS/active layer/PNDIT-F3N/Ag) was used with active layer thicknesses of approximately 120 nm, 300 nm, and 500 nm. A layer-by-layer (LBL) processing strategy was used to optimize the vertical phase separation of the ternary blend. In this approach, the PM6 donor was spin-coated onto the PEDOT:PSS layer, followed by spin coating of the acceptor blend (BTP-eC9:L8-BO-F). The photovoltaic performance of the devices was characterized by measuring current-voltage (J-V) curves under simulated AM 1.5 G illumination and external quantum efficiency (EQE) spectra. Space-charge-limited current (SCLC) measurements were used to determine hole and electron mobilities. Transient absorption (TA) spectroscopy was employed to investigate the charge recombination dynamics. Film-depth-dependent light absorption spectroscopy (FLAS) in combination with optical transfer-matrix models was used to investigate vertical phase segregation and exciton generation contours. Grazing incidence wide-angle X-ray scattering (GIWAXS) and atomic force microscopy (AFM) were used to investigate the morphology of the active layers. Steady-state photoluminescence (PL) and time-resolved photoluminescence (TRPL) measurements were used to investigate exciton diffusion. Energy loss analysis, including radiative and non-radiative losses, was also performed using electroluminescence (EL) data. As a comparison, a second set of experiments was conducted using the donor PM6 and acceptors Y6 and Y6-F, following similar methods.

The study achieved significant improvements in the efficiency of thick-film OSCs using a combination of material selection and processing techniques. The ternary blend OSCs (PM6:BTP-eC9:L8-BO-F) showed substantially higher PCEs compared to their binary counterparts (PM6:BTP-eC9), especially at larger thicknesses. The 300 nm thick ternary device achieved a remarkable PCE of 17.31% (certified value of 16.9%), with a Jsc of 28.36 mA cm⁻², Voc of 0.836 V, and FF of 73.0%. The 500 nm thick device still maintained a respectable PCE of 15.21%. The layer-by-layer (LBL) processing method further enhanced the performance. The LBL-processed 300 nm thick ternary device exhibited even higher PCE than the non-LBL processed counterpart, demonstrating the effectiveness of this strategy. Measurements of exciton diffusion lengths revealed that the blend of BTP-eC9 and L8-BO-F showed a significantly increased exciton diffusion length (47 nm) compared to the individual acceptors (36.6 nm and 44.4 nm, respectively). This enhancement is attributed to efficient Förster energy transfer between the two acceptors and increased crystallinity in the blend. Space charge limited current (SCLC) measurements showed that the LBL-processed ternary device had balanced hole and electron mobilities (μh = 12.33 × 10⁻⁴ and μe = 9.04 × 10⁻⁴ cm² V⁻¹ s⁻¹). The LBL processing resulted in a graded vertical phase separation with an acceptor-rich top layer, promoting efficient charge collection. Transient absorption spectroscopy showed that the LBL-processed devices had longer charge carrier lifetimes compared to the non-LBL devices. Energy loss analysis indicated that the ternary blends had reduced non-radiative recombination losses. The success was replicated using PM6, Y6, and Y6-F, verifying the methodology.

The findings demonstrate that a synergistic effect between enhanced exciton diffusion and vertically optimized phase separation is crucial for achieving high-efficiency thick-film OSCs. The enlarged exciton diffusion length in the mixed acceptor phase allows excitons generated throughout the thicker active layer to reach the donor-acceptor interface for efficient charge separation. The LBL processing technique creates a graded vertical phase separation that enhances charge transport and reduces recombination. The combination of these two improvements led to significant enhancements in both short-circuit current density (Jsc) and fill factor (FF). The results are impactful because they provide a pathway toward scalable, cost-effective manufacturing of high-efficiency OSCs by using thicker active layers that are easier to manufacture. The success with two different NFA combinations (BTP-eC9:L8-BO-F and Y6:Y6-F) indicates that the strategy is broadly applicable.

This research successfully demonstrated efficient thick-film organic solar cells with PCEs exceeding 16.9% (certified) even with 300 nm active layers. The key strategies involved the use of a ternary blend of donor and two non-fullerene acceptors with enhanced exciton diffusion length and the implementation of a layer-by-layer processing method to optimize vertical phase separation. Future work could explore the use of different donor materials or acceptors with even longer exciton diffusion lengths to further enhance efficiency. Investigating alternative processing techniques to control the morphology at larger thicknesses would also be beneficial. The generalizable nature of this approach could pave the way for mass production of high-efficiency organic solar cells.

While this study achieved significant advancements in thick-film OSCs, some limitations remain. The study primarily focuses on two specific combinations of donor and acceptor materials. The generalizability to other material systems needs further investigation. Although the LBL processing improved efficiency, the fabrication process is more complex than conventional spin coating. Further optimization and automation of the LBL technique might be necessary for large-scale manufacturing. The long-term stability of the thick-film devices over extended periods needs further assessment under diverse environmental conditions.