Multifunctional sulfonium-based treatment for perovskite solar cells with less than 1% efficiency loss over 4,500-h operational stability tests

Efficiency, stability and scalability are the most important factors on the route towards commercialization of perovskite solar cells (PSCs). While record power conversion efficiencies (PCEs) have reached 25.7% for single-junction PSCs and 32.5% for perovskite–silicon tandems, stability remains the most critical limitation for commercialization. Performance loss and degradation in PSCs are initiated at grain boundaries and interfaces where defects and mobile ions accumulate under stresses such as continuous illumination, humidity, and elevated temperature. Suppressing surface defects and inhibiting ion migration are therefore essential for long-term stability. Various passivation strategies using ammonium salts, small organic molecules, polymers and inorganic salts have been explored. Aprotic sulfonium-based molecules have shown promise but remain largely unexplored; their geometry and chemical robustness can enhance stability, yet prior aliphatic sulfonium salts suffer from poor solubility in low-polar solvents, limiting surface treatment use. In this work, the authors synthesize and introduce a dimethylphenethylsulfonium cation and its iodide salt (DMPESI) as a surface passivation agent for FAPbI3-based PSCs to address these challenges.

The paper situates the work within advances in PSC efficiency and scalability, citing certified PCEs up to 25.7% for single-junction and 32.5% for perovskite–silicon tandem devices, and scalable deposition strategies achieving >20% on mini-modules. Stability, however, lags and is often limited by grain boundary/interface defects and ionic motion. Prior passivation approaches include ammonium-based salts, small molecules, polymers, and inorganic agents. Aprotic sulfonium-based cations have emerged as promising stabilizers due to enhanced chemical and humidity resistance; examples include trimethylsulfonium-based 1D perovskites and butyl-dimethylsulfonium iodide additives that improved humidity stability. A key limitation of these aliphatic sulfonium salts is poor solubility in low-polar solvents (e.g., isopropanol, chloroform), hindering their surface treatment applications. This motivates the development of an aromatic sulfonium cation (DMPESI) expected to promote closer packing and robust, hydrophobic surface layers.

The study synthesizes dimethylphenethylsulfonium iodide (DMPESI) via methylation of 2-mercaptophenylethane followed by quaternization to yield the sulfonium iodide salt, with structural verification by 1H and 13C NMR. FAPbI3 perovskite powders are prepared from FAI and PbI2 in anhydrous solvents. Devices are fabricated in an n–i–p architecture: FTO/cp-TiO2/mp-TiO2/3D FAPbI3/DMPESI/spiro-OMeTAD (or PTAA)/Au (80 nm), with an active area of 0.16 cm2. The perovskite precursor (FAPbI3 1.8 M with MACl 0.63 M in DMF:DMSO 4:1) is spin-coated with antisolvent quench and annealed at 150 °C for 10 min. DMPESI is dissolved in chloroform at 1–10 mg ml−1 and spin-coated on the perovskite, followed by 100 °C drying. Hole transport layers (spiro-OMeTAD or PTAA) are prepared with standard dopants. Characterization includes contact angle measurements, SEM/AFM morphologies, XRD and GIWAXS for phase/crystallinity, and wide-field hyperspectral photoluminescence (PL) microscopy and time-resolved PL decays (including differential lifetime analysis and drift-diffusion simulations) to assess recombination and charge extraction. Solid-state MAS NMR (1H, 13C; including 1H–1H spin diffusion and 1H–13C CP) probes DMPESI–perovskite interactions and local environments. Scanning electron diffraction (SED) with annular dark-field reconstructions examines microstructure and intragranular phases. DFT calculations (adsorption/binding energies, solvation/water interaction energies, electronic structure and defect passivation) are performed for DMPESI and PEAI on PbI2- and FAI-terminated surfaces at varying coverages. ToF-SIMS depth profiling quantifies ion (I–) and Au migration in fresh vs aged devices. Stability testing comprises: shelf storage in ambient air (R.H. 20–40%); unencapsulated thermal aging at 60 °C (<30% R.H.); ISOS-T-1 thermal cycling (25–85 °C) on encapsulated devices; damp-heat (85 °C, 85% R.H.) on encapsulated devices; and long-term operational stability under continuous 1-sun equivalent illumination and maximum power point tracking (MPPT) in N2 flow at room temperature. JV, stabilized MPP, and light-intensity dependent VOC (ideality factor) measurements assess performance and recombination.

• Hydrophobic and chemically robust DMPESI: 1H NMR shows no H2O uptake after six months in ambient air; DFT indicates reduced water binding vs phenethylammonium (PEA) and a less polar electrostatic potential. • Film hydrophobicity and morphology: Water contact angle increases from 55.8° (reference) to ~90° for DMPESI-treated films across 1–10 mg ml−1, indicating a more hydrophobic surface. Moderate DMPESI concentrations (1–5 mg ml−1) yield uniform, smooth films; at 10 mg ml−1, aggregation and surface cracking appear. • Phase stability under ambient and humidity: Reference FAPbI3 films discolor within 10 days and convert to yellow phase under 85–95% R.H. in 24 h, while DMPESI-treated films (1–5 mg ml−1) retain the black phase. Films with 10 mg ml−1 show aggravated phase instability. • Light-soaking stability: Hyperspectral PL mapping under 1-sun shows the reference develops large PL intensity drops, blue shifts (~75 meV), and high-energy peaks (~2 eV) within 10 min, indicating phase/composition changes. DMPESI-treated films exhibit negligible changes in PL intensity, distribution, or spectra over the same period. • Atomistic interactions and passivation: DFT shows stronger dissociative adsorption/binding of DMPESI vs PEAI on PbI2-terminated surfaces (single-molecule adsorption energies: 2.09 eV vs 1.74 eV; half coverage: 1.93 eV vs 1.71 eV per molecule). On FAI-terminated surfaces, DMPESI forms a strong, dispersion-dominated protective layer (1.69 eV vs 1.28 eV for PEAI). DMPESI encapsulates mobile iodides, reduces interstitial iodine defects, and lowers surface trap states without significantly perturbing the FAI-terminated electronic structure. • Microstructure and interfacial chemistry: SED reveals superstructure reflections consistent with octahedral tilting that hinder δ-phase formation and suggests low-dimensional perovskite-like species at grain boundaries upon DMPESI treatment. Solid-state NMR (1H, 13C; CP and spin diffusion) indicates altered chemical environments and intimate proximity between DMPESI and FA, confirming interaction/reaction at the perovskite interface. • Optoelectronics and recombination: Differential lifetime analysis and simulations indicate >16× reduction in surface recombination for DMPESI-treated stacks, with optimal concentration at 3 mg ml−1 balancing suppression of interfacial recombination and efficient hole extraction. Higher concentrations (>3 mg ml−1) hinder charge extraction without further recombination reduction. • Device performance: With 3 mg ml−1 DMPESI, champion JV shows RS PCE 23.32% (JSC 25.53 mA cm−2, VOC 1.136 V, FF 0.805) and FS PCE 23.14% (JSC 25.39 mA cm−2, VOC 1.135 V, FF 0.803); stabilized MPP PCE 23.3%. Hysteresis is negligible. Light-intensity VOC yields ideality factor n reduced from 1.74 (control) to 1.38 (treated), indicating suppressed trap-assisted recombination; non-radiative FF losses reduced by ~2×. • Durability: Unencapsulated shelf storage (67 days, R.H. 20–40%) shows control loses ~60% PCE, while DMPESI-treated retains 94%; treated devices remain black by XRD. Unencapsulated at 60 °C (<30% R.H., 400 h): treated retains >96% PCE vs control <85%. Encapsulated thermal cycling (25–85 °C, ISOS-T-1): treated retains >98% of initial PCE. Encapsulated damp-heat (85 °C, 85% R.H., >1,000–1,050 h): <5% PCE loss. Long-term MPPT in N2 at r.t.: treated devices show <1% PCE drop over >4,500 h; linear extrapolation indicates multi-year lifetime (reported theoretical T80 over nine years; figure extrapolation T90 over nine years). • Ion migration suppression: ToF-SIMS after ageing reveals strong I− and Au diffusion into HTL and perovskite for controls, but negligible diffusion in DMPESI-treated devices, correlating with preserved VOC/FF and stability.

The study addresses the core stability challenge in FAPbI3 PSCs by targeting grain boundary and surface defects, as well as ion migration pathways. DMPESI’s aprotic sulfonium cation forms a strongly interacting, hydrophobic interfacial layer on both PbI2- and FAI-terminated surfaces, effectively reducing surface trap states and binding or sequestering mobile iodide. This dual function mitigates moisture ingress, inhibits light/thermal-induced phase transformations, and suppresses non-radiative recombination, particularly at the perovskite/HTL interface. Microscopically, the presence of octahedral tilting and intragranular low-dimensional species at grain boundaries synergize with surface passivation to stabilize the black phase. Optoelectronic analysis shows that at optimal coverage (≈3 mg ml−1), surface recombination is substantially reduced without impeding charge extraction, translating into higher VOC, FF, negligible hysteresis, and 23.3% stabilized MPP PCE. Operationally, the treatment eliminates burn-in and maintains performance over thousands of hours under 1-sun MPPT, while also meeting stringent ISOS thermal cycling and damp-heat conditions with minimal efficiency loss. These results demonstrate that tailored sulfonium-based surface treatments can deliver the efficiency–stability combination needed for industrially relevant perovskite photovoltaics and highlight a mechanistic route—strong interfacial binding, hydrophobic capping, and ion migration suppression—to durable device operation.

A single-step surface treatment with the aromatic sulfonium salt DMPESI stabilizes the black phase of FAPbI3 and dramatically enhances PSC durability and performance. Strong interfacial interactions and a hydrophobic overlayer passivate surface states, reduce trap-assisted recombination, and suppress migration of iodide and metal ions, while intragranular species at grain boundaries further bolster phase stability. Devices achieve 23.3% stabilized PCE and exhibit exceptional stability: <1% loss after >4,500 h MPPT, >98% retention under ISOS thermal cycling, and <5% loss after >1,000 h damp heat, alongside robust shelf and thermal aging behavior. This work opens a pathway for deploying multifunctional sulfonium-based molecules in perovskite optoelectronics and suggests further exploration of molecular design, coverage control, and interface engineering to optimize charge extraction while maintaining maximal stability. Future work could extend to large-area devices and modules, varied transport layers, and accelerated lifetime testing under outdoor-representative conditions.

Performance depends on treatment coverage: excessive DMPESI (e.g., 10 mg ml−1) aggregates, introduces cracks, can form overlayers that hinder hole extraction, and aggravates phase instability. Even at full coverage, DFT indicates overlayer formation where not all molecules bind the perovskite surface, potentially limiting full passivation and charge extraction. Operational stability measurements were primarily conducted under N2 flow at room temperature (for MPPT), and device areas were small (0.16 cm2), although ISOS thermal cycling and damp-heat tests were performed on encapsulated devices. Optimization is needed to balance interfacial passivation with transport, as concentrations >3 mg ml−1 reduce charge extraction efficiency.