Keyou Yan

Solvent optimization in organic solar cells (OSCs) remains largely empirical, requiring repeated device fabrication and characterization across a broad formulation space. Here, we introduce a high‐throughput workflow to decipher a figure of merit (FoM) for solvent optimization of non‐fullerene acceptor (NFA)‐based OSCs. Using D18:L8BO as a representative system, we prepare a diverse solvent formulation via high‐throughput experimentation and in situ optical characterization during film formation. Machine learning analysis unveils the initial donor‐to‐acceptor absorption ratio during deposition (I_D/A) as a key FoM, which enables the efficient optimization of solvents in OSCs. Comprehensive characterization identifies the photophysical and morphological properties associated with the optimized formulation. Evaluation across several polymer:Y‐series NFA systems further supports I_D/A as a FoM for solvent optimization. This study establishes an experimentally accessible strategy that integrates the exploration of broad solvent formulation spaces with targeted device validation into a targeted, data‐driven workflow, offering a practical route for accelerating OSC development.

Metal halide perovskites (MHPs) have emerged as promising photocatalysts owing to their exceptional optoelectronic properties, yet suffer from rapid charge recombination, sluggish surface kinetics, and instability. Facet engineering, the controlled exposure of specific crystallographic planes, offers an atomic‐level strategy to tailor surface properties and address these challenges. This review systematically summarizes recent advances in the facet engineering of MHP photocatalysts. It begins with the fundamental principles of how facets govern light harvesting, charge separation, surface reactions, and stability. Key synthetic strategies for facet control are then outlined, followed by a critical discussion on enhanced photocatalytic performance in CO ₂ reduction, H ₂ evolution, and organic transformations. The synergy of facet engineering with heterojunctions, doping, and defect engineering is also highlighted. Finally, challenges and future perspectives are presented, focusing on stability, precise facet control, high‐index facets, lead‐free perovskite systems, and advanced in situ characterization. This review aims to guide the rational design of efficient and stable MHP photocatalysts through facet manipulation for solar energy conversion and environmental remediation.

Phase separation and structural damage induced by ion migration represent primary bottleneck in blue perovskite light‐emitting diodes (PeLEDs). Herein, we developed heterocyclic diammonium cations, namely 3/4‐aminopiperidine (3AP ²⁺ /4AP ²⁺ ) dihydrochloride, for rigid perovskites, that effectively mitigates the degradation pathway. Compared to conventional phenylethylammonium (PEA ⁺ ), these piperidinium‐based dications promote the formation of a compact and mechanically reinforced perovskite film. This structural enhancement is quantitatively demonstrated through a remarkable increase in Young's modulus, with values reaching 24.5/20.1 GPa for 3AP/4AP‐incorporated films, representing 1.7/1.4‐fold improvements over PEA‐based counterparts, respectively. The observed improvement in mechanical properties is further corroborated by molecular dynamics simulations. The strengthened framework substantially suppresses ion migration and preserves structural integrity, leading to exceptional spectral stability under both optical and electrical excitation. Consequently, we demonstrate high‐performance sky‐blue (488 nm) and pure‐blue (475 nm) PeLEDs with peak external quantum efficiencies (EQEs) of 18.8% and 9.8%, respectively. This work establishes molecular rigidification as an effective approach for developing high‐performance and operationally stable blue perovskite optoelectronics.

All‐inorganic perovskite solar cells are photovoltaic research hotspots due to their excellent thermal stability. To boost efficiency, ZnO is used as an electron transport material for its high mobility and energy level matching with inorganic perovskites. However, abundant surface defects on CsPbI ₂ Br trigger undesirable interfacial reactions with ZnO, seriously limiting device performance and long‐term stability. Herein, a two‐birds‐with‐one‐stone interfacial modifier based on HTPT (4‐hydroxy‐6‐(trifluoromethyl)pyrimidine‐2‐thiol) is utilized. S, O, and N heteroatoms in HTPT passivate defects via multi‐site coordination, reducing reactive sites. The steric effect of CF ₃ partially reduces direct contact between perovskite and ZnO, and various interfacial effects, including surface energy variation, interfacial dipole modulation, wettability, and hydrophobicity, jointly regulate the overall interface properties, constructing an effective strategy for suppressing interfacial interdiffusion and slowing interfacial degradation rates. Modified single‐junction p‐i‐n devices reach a record 17.34% efficiency with 1.32 V open‐circuit voltage and robust stability (T ₈₀ = 864 h at 85°C, T ₉₂ = 1000 h under maximum power point tracking (MPPT) at 30°C). P‐i‐n CsPbI ₂ Br/PM6:eC9:PC61BM two‐terminal tandem cells were fabricated. The fabricated CsPbI ₂ Br/organic tandem cells yield top efficiencies of 24.25% (0.08 cm ² ) and 22.81% (1.02 cm ² ) among p‐i‐n all‐inorganic perovskite/organic tandems and maintain decent thermal and MPPT stability.

Scalable perovskite photovoltaics are limited by the coupled challenges of unstable SnO ₂ colloids and defect‐rich buried interfaces, which together undermine film uniformity and charge extraction. While existing strategies typically target either colloidal stabilization or interface passivation alone, their single‐function nature restricts effectiveness for large‐area processing. Here, we introduce potassium dihydrogen phosphate (KH ₂ PO ₄ ) as a dual‐functional modifier that produces a molecular disaggregation–bridging effect (MDBE). Dihydrogen phosphate anions regulate the colloidal organization of commercial SnO ₂ dispersions, contributing to smoother and more uniform electron‐transport layers, while simultaneously establishing phosphate‐mediated interfacial coordination at the buried SnO ₂ /perovskite interface. This integrated regulation suppresses trap‐assisted recombination and promotes uniform perovskite crystallization. Consequently, perovskite devices fabricated via blade coating in combination with vacuum evaporation reach a PCE of 25.34%, and 5 × 5 cm ² modules (12 cm ² active area) deliver 23.16% with only 8% loss upon scaling. MDBE thereby provides a unified and cost‐effective interface‐engineering strategy that overcomes the limitations of conventional single‐function additives, enabling reliable and high‐performance large‐area perovskite photovoltaics.

Experience‐driven optimization has been pivotal in advancing organic photovoltaics, where chemically intuitive and stepwise tuning of processing parameters can yield high‐performance organic solar cells (OSCs). However, locating optimal regions under high‐dimensional, coupled, and multi‐objective constraints remains challenging across diverse scenarios. Here, we introduce a closed‐loop workflow base on multi‐objective Bayesian optimization for efficient optimization of OSCs. Applied to a quaternary system comprising the polymer donors PM6 and D18 and the non‐fullerene acceptors BTP‐eC9 and L8BO, the workflow navigates an eight‐dimensional space defined by composition and fabrication variables, encompassing 2.2 × 10 ¹⁴ possible combinations. Within five active‐learning cycles, the search converges to a high‐performance region, achieving the efficiency exceeding 20%. The generalizability of this workflow is further demonstrated across different materials and fabrication methods, guiding the optimization toward high‐performance regions within few iterations. This work establishes an efficient and scalable closed‐loop framework for rapidly identifying favorable processing windows in organic photovoltaics.

Formamidinium–cesium (FACs) perovskite is thermally stable, but the rapid crystallization process generates bulk defects and residual strains. We introduce a multifunctional additive, 3‐(Trifluoromethyl)aniline hydrochloride (3TFACl), to effectively regulate the crystallization process to improve stability and scalability of perovskite solar cells (PSCs) through multiple interactions with perovskite components. Crystallization‐defect‐strain engineering is synergistically integrated during film deposition. The champion power conversion efficiency (PCE) of p–i–n PSCs and 12 cm ² module reach 26.76% (certified 26.12%) and 22.77%, respectively. The unencapsulated devices still maintain more than 94.36% of the initial efficiency after 1500 h maximum power point tracking (MPPT) (ISOS‐L‐1I protocol) and 80.5% initial efficiency after 1000 h under 85°C (ISOS‐D‐2I protocol). This work proposes a molecular design strategy with a simple structure but high synergy to achieve efficient, stable, and scalable perovskite photovoltaic devices.

The solid electrolyte interface (SEI) of Li metal batteries (LMBs) generated by traditional carbonate electrolytes is extremely unstable and prone to repeated rupture, resulting in a large amount of Li dendrites and dead Li, which seriously affects the cycling stability. In this work, we introduced trifluoromethanesulfonic acid trimethylsilyl ester (PMS) and LiNO3 as dual synergistic additives, and successfully constructed a Li3N/Li2SO3/LiF-rich SEI on Li metal anode. By utilizing the high binding energy between PMS and NO3-, the crystallization/ionization equilibrium of LiNO3 is disrupted, increasing the solubility of LiNO3 by nearly 4 times (1.2wt%) in carbonate electrolyte. In addition, LiNO3 and PMS both have low LUMO energy levels and can decompose on SEI to generate Li3N, Li2SO3 and LiF. Under the synergistic effect of Li3N with excellent rate performance and Li2SO3/LiF with high interfacial energy, the Li//Li symmetric battery can stably cycle for over 700 h at 1.0 mA cm-2 and 1.0 mAh cm-2 conditions. Li//LFP (10.41 mg cm-2) and Li//NCM811 (10.82 mg cm-2) full batteries can stably cycle for 800 cycles at 1 C with a capacity retention rate of 70% and for over 200 cycles at 0.4 C with a capacity retention rate of 80%, respectively.