Graded bulk-heterojunction enables 17% binary organic solar cells via nonhalogenated open air coating

The study addresses limitations of conventional bulk heterojunction (BHJ) organic solar cells, including complex morphology control, reliance on toxic halogenated solvents, degraded performance in thick films due to transport bottlenecks and traps, and morphology sensitivity when moving from spin to blade coating. Inspired by gradient junction concepts in inorganic photovoltaics, the authors hypothesize that a graded bulk-heterojunction (G-BHJ) with controlled vertical donor–acceptor composition and crystallinity gradients can improve charge separation and transport, reduce recombination, enable thick active layers, and support eco-friendly, scalable processing. They propose sequential deposition (SD) using solvent engineering to form G-BHJ films with tunable interdiffusion, aiming to deliver high PCEs with nonhalogenated solvents and robust blade-coated devices.

Prior work shows Y6-series nonfullerene acceptors have pushed single-junction BHJ OSC efficiencies beyond 17–18%, but with significant constraints: narrow solvent windows often involving chlorinated solvents, delicate morphology tuning via additives/annealing, and poor thickness tolerance. Sequential deposition has been successful with fullerene acceptors using orthogonal solvents to achieve graded vertical distributions (p-i-n-like), but application to NFA systems is challenging due to similar solubilities of donor and acceptor, limited orthogonal solvent pairs, and insufficient intermixing. Although SD-based high-performance NFA OSCs have been reported, a clear quantitative picture of the vertical composition/crystallinity and a rational solvent selection guideline have been lacking. Additionally, scale-up via blade coating often induces excessive NFA aggregation from slow drying, leading to oversized domains unless complex post-treatments are used. This work positions G-BHJ via SD, with quantitative depth profiling and processing-solvent rules, to overcome these gaps.

- Materials/system: Polymer donor PM6 and NFA BTP-eC9 in conventional ITO/PEDOT:PSS/active layer/PFN-Br/Ag devices. - G-BHJ formation by sequential deposition: First deposit neat PM6 layer, then deposit neat BTP-eC9 layer using tailored upper-layer solvents. Solvent selection balances solubility and boiling point to control interdiffusion time and penetration rate; two effective pathways demonstrated: o-xylene (XY, high boiling, low solubility) and chloroform (CF, low boiling, high solubility). Additive 1,8-diiodooctane (DIO) used to tune interdiffusion/crystallization. - Spin-coated devices: Optimize layer thicknesses and DIO content (e.g., PM6 ~70 nm, BTP-eC9 ~55 nm, 0.5% DIO in XY), including CF-processed counterparts for comparison; fabricate BHJ references. - Blade coating (open air): Air-knife-assisted blade coating from XY with controlled substrate temperature (60 °C), nitrogen knife quenching (40 m s−1), and DIO optimization; also CF-based blade-coated G-BHJ as control. Active area 0.04 cm2. - Thickness study: Fabricate G-BHJ devices with active layers from ~120 to 500 nm by adjusting concentrations and spin speeds; fabricate BHJ controls. - Characterization of vertical morphology: Depth-profiling XPS (DP-XPS) quantifying polymer wt% vs depth using O/F ratios; ToF-SIMS profiling (CN signal) cross-validation. - Crystallinity/orientation: Angle-dependent GI-XRD (incident angles 0.15°, 0.20°, 0.25°) to probe near-surface and bulk; extract lamellar and pi–pi crystalline coherence lengths (CCL) via Scherrer analysis. - Device/electrical metrics: J–V under AM1.5G; EQE and Jsc integration; photocurrent vs Veff (eta_diss and eta_coll); VOC vs ln(light intensity) for monomolecular recombination; Jsc vs P for bimolecular recombination exponent; SCLC hole/electron mobilities using single-carrier diodes. - Morphology imaging: AFM (height/phase, RMS); TEM for bulk phase separation. - In situ film formation: Time-resolved UV–vis absorption during annealing for BHJ and G-BHJ blade-coated films to extract aggregation rates and peak evolution for donor/acceptor.

- High efficiencies via SD G-BHJ: - XY-processed spin-coated G-BHJ: PCE 17.48%, Voc 0.840 V, Jsc 26.65 mA cm−2, FF 0.781; EQE peak ~88% at ~650 nm; integrated Jsc 26.04 mA cm−2. - CF-processed spin-coated G-BHJ: PCE 17.54% (higher than CF-BHJ 17.08%). - XY-processed BHJ (reference): PCE 16.41%, Jsc 25.75 mA cm−2, FF 0.760. - Open-air blade-coated devices (nonhalogenated XY): - G-BHJ: PCE 16.77%, Voc 0.836 V, Jsc 26.26 mA cm−2, FF 0.764; integrated Jsc 25.30 mA cm−2. - BHJ: PCE 15.87%, Voc 0.835 V, Jsc 25.24 mA cm−2, FF 0.753. - Thickness tolerance (G-BHJ, XY, spin-coated): - 120 nm: 17.48% (Voc 0.840 V, Jsc 26.65, FF 0.781). - 300 nm: 16.25% (Voc 0.830 V, Jsc 26.89, FF 0.728). - 400 nm: 15.12% (Voc 0.823 V, Jsc 27.42, FF 0.670). - 500 nm: 14.37% (Voc 0.821 V, Jsc 27.31, FF 0.641). Among highest thick-film OSCs reported. - FF degradation with thickness is milder in G-BHJ than BHJ (BHJ FF: 0.76 at ~110 nm to 0.58 at 500 nm; G-BHJ FF: 0.78 at 120 nm to 0.64 at 500 nm). - Vertical composition gradients (DP-XPS/ToF-SIMS): - G-BHJ shows acceptor-enriched top (region I) and donor-enriched bottom (region III) with a middle region (II). Without DIO: top ~15.8 wt% polymer, bottom ~75.5 wt%; with DIO: more polymer in bulk region II and bottom ~66.4 wt% polymer, indicating DIO assists acceptor downshift and improved intermixing. Thick (500 nm) G-BHJ shows stronger gradient (top ~16 wt%, bottom ~90 wt% polymer). BHJ with additive shows undesired higher donor content near anode. - Crystallinity gradients (angle-dependent GI-XRD): - Both films maintain face-on orientation. G-BHJ enhances PM6 lamellar CCL vs BHJ (PM6 (100) CCL ~113 Å vs ~92.6 Å at ~0.20°), indicating improved polymer ordering; similar overall (010) pi–pi CCLs (~19.9–20.6 Å). - At surface (0.15°), G-BHJ shows larger BTP-eC9 lamellar CCL (~62.8 Å) than BHJ (~54.3 Å), implying more ordered acceptor near top; PM6 top-layer crystal sizes similar. - With increased angle (0.25°), PM6 lamellar CCL increases more in G-BHJ (from ~80.7 Å to ~102 Å) than in BHJ (to ~94.2 Å), evidencing stronger PM6 crystallinity gradient (weaker at top, stronger at bottom) in G-BHJ; BTP-eC9 shows decreasing crystal size from top to bottom. - DIO further prolongs CCLs of both components and balances top-to-bottom crystallization. - Charge generation/transport: - Higher eta_diss (96.03%) and improved eta_coll in G-BHJ; reduced bimolecular recombination (Jsc vs P exponent S ~0.98 vs BHJ 0.95) and similar trap-assisted recombination (VOC vs lnP slopes ~1.21–1.24 kT/q). - PL quenching stronger in G-BHJ: vs neat PM6, BHJ 97.6%, G-BHJ 99%; vs neat BTP-eC9, BHJ 94.5%, G-BHJ 98.1%. - SCLC mobilities: BHJ mu_e 6.32e-4, mu_h 4.15e-4 cm2 V−1 s−1 (mu_e/mu_h ~1.52); G-BHJ mu_e 5.34e-4, mu_h 5.11e-4 (ratio ~1.05), indicating more balanced transport and fewer isolated domains, contributing to higher FF. - Film formation kinetics (in situ UV–vis, blade-coated): - G-BHJ films solidify faster (8.6 s) than BHJ (12.9 s). Polymer pre-aggregation stage observed in BHJ is suppressed in G-BHJ. - Aggregation rates (s−1): BHJ acceptor 0.083, donor 0.020; G-BHJ acceptor 0.281, donor 0.072. Faster, more balanced aggregation in G-BHJ mitigates excessive NFA aggregation and leads to favorable phase separation. - Solvent selection guideline: Achieve G-BHJ by balancing upper-layer solvent solubility and boiling point to tune penetration rate and interdiffusion time (e.g., XY: low solubility/high boiling; CF: high solubility/low boiling). Additives provide an extra tuning dimension. - Morphology (AFM/TEM): G-BHJ surface shows granular aggregates without DIO and optimized mixed thread-like/granular features with DIO; RMS roughness BHJ 1.68 nm, G-BHJ without DIO 0.91 nm, optimized G-BHJ 1.62 nm; blade-coated G-BHJ RMS ~1.14 nm, indicating suppressed aggregation vs BHJ. TEM indicates similar bulk-scale D–A phase separation across films.

The results validate that a controlled vertical composition and crystallinity gradient (acceptor-rich top/electron pathways, donor-rich bottom/hole pathways) formed via sequential deposition improves charge generation, collection, and transport relative to BHJ. The G-BHJ morphology alleviates the typical Jsc–FF trade-off by simultaneously enhancing exciton dissociation, reducing recombination, and balancing carrier mobilities. This morphology is robust across a wide thickness range, enabling high FFs and PCEs even at 300–500 nm, which is important for scalable, thickness-tolerant manufacturing. The demonstrated open-air blade-coated G-BHJ using a nonhalogenated solvent attains record performance, with in situ spectroscopy indicating faster and more balanced donor/acceptor crystallization that suppresses excessive acceptor aggregation typical in slow-drying BHJ films. The solvent selection rule—balancing solubility and volatility to tune interdiffusion—together with additive engineering provides a rational framework to design G-BHJ processing across material systems and coating methods, offering a practical pathway toward eco-friendly and manufacturable high-efficiency OSCs.

This work introduces a solvent-guided sequential deposition strategy to form graded bulk-heterojunctions with quantified vertical composition and crystallinity gradients, achieving 17.5% class spin-coated and 16.8% open-air blade-coated PCEs using a nonhalogenated solvent, and maintaining high efficiencies up to 500 nm thickness. Depth-profiling XPS, ToF-SIMS, and angle-dependent GI-XRD reveal donor-enriched bottom and acceptor-enriched top with complementary crystallinity gradients that yield balanced transport and reduced recombination. A simple solvent selection guideline (balancing solubility and boiling point) plus additive tuning generalizes the approach. Future work should: expand to diverse donor–acceptor systems with tailored “green” solvent pairs; design materials compatible with solvent-driven interdiffusion; employ in situ GI-XRD/electron microscopy to further resolve film-formation dynamics; and scale up to larger-area modules optimizing blade/slot-die coating parameters to translate G-BHJ advantages to manufacturing.

- Scope is demonstrated primarily on PM6:BTP-eC9 with supporting examples from two additional systems; broader generality across diverse materials remains to be established. - Device areas are small (0.04 cm2); scalability to large-area modules and long-term operational stability were not evaluated here. - The solvent selection rule, while practical, may require system-specific optimization of solvent/additive concentrations and processing conditions. - Depth-profiling and grazing-incidence techniques provide indirect, ensemble-averaged structural information; nanoscale 3D morphology and potential sputter-induced artifacts are not fully addressed.