Translate this page into:
Performance enhancement of perovskite solar cells for energy storage applications: Materials and device engineering
* Corresponding author: E-mail address: jxren@cust.edu.cn (J. Ren)
-
Received: ,
Accepted: ,
Abstract
Developing highly stable perovskite solar cells (PSCs) is an important prerequisite for their future use in energy-storage-related applications. However, interfacial degradation driven by ion migration and encapsulation failure under thermal and humid stress remain a major obstacle to durable PSC operation. In this work, a coordinated materials-and-device engineering strategy is developed to enhance PSC stability through the integration of silane self-assembly interfacial passivation, an SnO2 buffer layer, low-temperature dynamic thermocompression encapsulation, and fluorosilicone edge protection. (3-Aminopropyl)triethoxysilane (APTES) treatment reduces interfacial defect density and suppresses ion-migration pathways, while the SnO2 nanolayer improves energy-level alignment and alleviates interfacial strain. The dynamic encapsulation process further improves sealing integrity and thermomechanical compatibility, and the fluorosilicone overcoat enhances hydrophobicity and edge-level environmental protection. X-ray photoelectron spectroscopy (XPS), ultraviolet photoelectron spectroscopy (UPS), and grazing-incidence X-ray diffraction (GI-XRD) analyses support the proposed interfacial coordination, work-function regulation, and lattice-strain mitigation effects. Under accelerated damp-heat aging at 85 °C and 85% relative humidity (RH), the optimized PSC devices retain over 92.5% of their initial power conversion efficiency after 1000 h, with a reduced trap-state density of 2.1 × 1016 cm-3. These results demonstrate that coordinated interface and encapsulation engineering can substantially improve the environmental durability of PSCs under application-relevant stress conditions and establish a practical device-level framework for stability enhancement.
Keywords
Dynamic thermocompression encapsulation
Energy-storage-oriented applications
Environmental stability
Materials and device engineering
Perovskite solar cells
Silane interfacial passivation

1. Introduction
Perovskite solar cells have become a cutting-edge research direction in the photovoltaic field due to their excellent light absorption coefficient and adjustable bandgap characteristics [1,2]. Perovskite solar cells (PSCs) combine strong visible-light absorption, tunable bandgaps, and low-temperature solution processability, making them highly attractive photovoltaic devices for next-generation lightweight energy technologies [3,4]. For any future use of PSCs in application scenarios that may involve intermittent operation, environmental exposure, or coupling with downstream power-management/storage units, device stability remains a fundamental prerequisite. In particular, two challenges are especially critical: interfacial degradation associated with ion migration and defect accumulation, and encapsulation failure arising from moisture/oxygen ingress and thermomechanical mismatch under prolonged thermal-humidity stress [5,6]. These coupled failure modes can progressively deteriorate charge transport, sealing integrity, and overall device durability. Therefore, it is important to develop coordinated materials and device engineering strategies that combine interfacial chemical stabilization with encapsulation-level mechanical and environmental protection under application-relevant conditions.
The performance evolution of perovskite photovoltaic devices within energy storage systems exhibits intricate multi-field coupling behaviors [7,8]. At charge transfer interfaces, ion migration and lattice strain collectively establish a dynamic equilibrium [9,10]. Under persistent illumination and thermal cycling, this metastable structure undergoes progressive evolution, thereby inducing periodic reconstruction of carrier transport pathways. The chemical potential gradient across interfacial regions drives the selective diffusion of metal cations and halide anions, generating electron trap states with diverse depth distributions near the band edges [11,12]. The response of the device encapsulation system to thermomechanical stress manifests pronounced time-dependent characteristics. Under cyclic thermal loading, the organic–inorganic heterogeneous interface experiences stress concentration, which initiates a synergistic failure mechanism involving interfacial debonding and crack propagation [13,14]. Notably, this electrochemically–mechanically coupled degradation process exhibits self-accelerating behavior: interfacial ion migration reshapes the local stress distribution, while mechanical strain in turn modulates the potential energy barrier for ion migration, thereby establishing a positive feedback loop [15,16]. Current research frameworks have not yet fully elucidated the governing mechanisms underlying this multi-scale coupling effect, and systematic synergistic optimization criteria remain lacking for material selection and structural design [17,18]. This gap constrains the long-term operational reliability of devices under realistic service conditions. Therefore, establishing a novel interface–encapsulation synergistic design theory is urgently required to overcome the existing performance bottlenecks [19,20]. To realize this objective, holistic optimization across the entire hierarchy—from atomic-scale chemical bonding modulation to macro-scale stress field management—is indispensable, which imposes more stringent demands on both advanced material characterization techniques and practical device engineering.
In response to ion migration and defect accumulation at perovskite interfaces, a variety of passivation strategies have been developed. For example, phenethylammonium iodide (PEAI) has been widely used for post-treatment of perovskite surfaces, where the formation of a two-dimensional perovskite capping layer can substantially reduce defect-state density and improve interfacial stability [21,22]. In a related study, Tsai H. and co-workers demonstrated that the instability of perovskite solid-state radiation detectors under electric-field stress is closely associated with ambient humidity, and that the introduction of a fluorinated PEAI blocking layer can effectively suppress voltage-induced degradation and improve device stability [23]. Nevertheless, such organic passivation strategies often remain sensitive to humidity and may face challenges in achieving uniform coverage over large areas. In addition to molecular passivation, NiOx-based interfacial engineering has also been explored to improve charge extraction and energy-level alignment because of the high work function of NiOx [24,25]. However, the lattice mismatch between NiOx and perovskite can still induce interfacial stress concentration, which limits its effectiveness in simultaneously maintaining structural and electronic stability. Beyond interface regulation, encapsulation engineering plays an equally important role in improving PSC durability under realistic operating conditions. Traditional encapsulation approaches, such as glass–metal solder sealing, can provide strong barrier performance, but their rigid structures are vulnerable to failure caused by thermally induced stress accumulation [26,27]. Although epoxy-based encapsulation offers a certain degree of flexibility, curing shrinkage can generate internal stress and induce microcracks in the encapsulation layer [28]. These limitations indicate that improving PSC stability requires not only effective defect passivation at the interface, but also mechanically adaptive and environmentally robust encapsulation strategies. Accordingly, interface molecular engineering and encapsulation-material innovation have emerged as important routes for enhancing PSC stability. Amino-functionalized silane molecules, for example, can passivate dangling bonds through coordination with Pb2+ on the perovskite surface, although the mechanical robustness of such single-layer passivation remains limited [29,30]. In parallel, fluoropolymer-based surface engineering has been investigated to improve coating adhesion and environmental resistance, highlighting the importance of mechanically robust protective layers for long-term operation [31,32]. Meanwhile, PSCs have attracted increasing interest for future coupling with energy-storage systems, in which operational stability becomes even more critical [33,34]. Recent studies have therefore further emphasized the importance of advanced materials and device-engineering strategies, including defect passivation, functional-layer design, and integrated stabilization approaches, for simultaneously improving PSC efficiency and durability [35].
These studies collectively highlight the importance of combining materials design and device-level engineering strategies to improve the performance and stability of perovskite solar cells. However, most previous reports have focused on a single functional layer or a single degradation pathway, such as interfacial passivation, transport-layer regulation, or encapsulation optimization. As a result, relatively few studies have considered how interfacial chemical compatibility and encapsulation-level mechanical/environmental protection can be coordinated within one stabilization framework. In particular, it remains challenging for single-route strategies to simultaneously suppress interface-related defects and improve encapsulation adaptability under thermal-humidity stress. In this work, silane self-assembly passivation is integrated with SnO2 buffer-layer regulation, low-temperature dynamic thermocompression encapsulation, and fluorosilicone edge protection to establish a coordinated materials-and-device engineering strategy. Rather than claiming each individual component as entirely unprecedented, the present study emphasizes their synergistic integration to simultaneously address interfacial ionic degradation and encapsulation-related thermomechanical failure under energy-storage-relevant operating conditions.
As summarized in Table 1, representative previous studies have typically improved PSC stability through one dominant route, such as molecular passivation, transport-layer optimization, or encapsulation design. By comparison, the present work seeks to bridge these previously separated stabilization routes through a coordinated interface–encapsulation engineering strategy. Specifically, silane self-assembly is used for defect passivation, an SnO₂ nanolayer is introduced for transport and stress buffering, low-temperature dynamic thermocompression is employed to improve encapsulation integrity, and fluorosilicone edge protection is added to enhance resistance to moisture and thermal stress. This comparison clarifies that the novelty of the present study lies in the synergistic integration of these functions, rather than in presenting each module as a completely unprecedented concept.
| Representative study | Main strategy | Primary benefit | Limitation relative to the present work |
|---|---|---|---|
| PEAI-based surface/interface passivation [21,22] | Interface passivation only | Reduces surface defects and improves carrier extraction | Mainly focuses on interfacial defect suppression; limited integration with encapsulation-level protection and thermomechanical stabilization |
| NiOx-based transport/interface engineering [24,25] | Transport-layer/interface regulation | Improves band alignment and charge transport | May introduce lattice/interface stress; does not address encapsulation durability or edge sealing |
| Low-temperature adhesive encapsulation [30] | Encapsulation only | Avoids thermal damage during sealing and improves device protection | Focuses on encapsulation performance, but does not simultaneously regulate interface defects or buffer-layer matching |
| Silane-based passivation strategies [31,32] | Molecular interfacial chemical passivation | Enhances defect passivation and device stability | Primarily addresses interface chemistry, without coordinated encapsulation and environmental barrier design |
| Fluoropolymer/fluorosilicone coating approaches [35] | Protective coating/adhesion enhancement | Improves hydrophobicity and coating adhesion | Mainly targets coating adhesion and barrier performance, without direct integration with interface passivation and dynamic encapsulation |
| This work | Integrated interface + buffer + encapsulation + edge protection | Combines interfacial defect suppression, buffer-layer regulation, low-temperature dynamic encapsulation, and edge protection for improved PSC durability | Focuses on materials/device-level stabilization; future system-level validation with coupled storage operation is still needed |
In this study, we focus on materials-and-device engineering strategies for improving the stability of PSCs under conditions relevant to future energy-storage-related applications. Although interfacial passivation, transport-layer optimization, and encapsulation design have been extensively investigated, these functions are often optimized as separate stabilization modules, leaving the coupled relationship between interfacial ionic degradation and encapsulation-related thermomechanical failure insufficiently addressed. Here, we develop a coordinated interface–buffer–encapsulation–edge-protection framework that integrates silane self-assembly passivation, an SnO₂ interfacial buffer layer, low-temperature dynamic thermocompression encapsulation, and fluorosilicone edge protection. Rather than claiming these individual components as entirely unprecedented, the novelty of this work lies in their synergistic integration to simultaneously enhance interfacial chemical stabilization, energy-level regulation, stress buffering, encapsulation integrity, and environmental protection. This strategy is designed to suppress moisture and oxygen ingress, reduce edge leakage, accommodate thermal-mechanical mismatch, and mitigate ion-migration-induced degradation within a unified device-level architecture. We further clarify that the present study does not include fully coupled PSC–battery operation, maximum power point tracking (MPPT) or direct current–direct current (DC–DC) converter validation, charge/discharge cycling, or grid-relevant testing. Instead, it establishes a stability-enhancing materials-and-device framework that may provide a practical basis for future PSC deployment in energy-storage-related application scenarios.
2. Materials and Methods
Figure 1 illustrates the overall materials-and-device engineering strategy developed to enhance PSC stability, spanning interfacial regulation, buffer-layer design, encapsulation, and edge protection. The perovskite absorber is modified with a silane self-assembled monolayer to reduce interfacial defects and suppress ion-migration pathways. The SnO₂ nano buffer layer contributes to energy-level alignment and stress buffering, thereby facilitating charge extraction and alleviating mechanical mismatch. Low-temperature dynamic thermocompression encapsulation improves interfacial filling and sealing integrity, while its gradient-modulus structure helps dissipate thermomechanical stress. In addition, the outer fluorosilicone copolymer coating provides hydrophobic protection and improves edge stability under thermal-humidity conditions. Together, these functional components establish a coordinated device-level stabilization framework that links interfacial chemical regulation with encapsulation-level mechanical and environmental protection across multiple length scales.

2.1. Silane self-assembly molecules to modify the interface
Amino-functionalized silane self-assembly molecules are added to the interface between perovskite and electrode, and a monolayer is formed by the solution method to achieve interfacial ionic bond compensation, inhibit the formation of migration channels, and improve interfacial electronic selectivity.
2.1.1. Solution method preparation of amino-functionalized silane monolayer and interfacial bonding regulation
(3-aminopropyl)triethoxysilane (APTES) is used as a self-assembly molecular precursor, and an isopropanol solution of appropriate concentration is prepared. After surface activation treatment, the perovskite film is immersed in the silane solution to form a dense monolayer. The silane molecules specifically bond with the defect sites on the perovskite surface through a hydrolysis condensation reaction. The process follows the Langmuir adsorption model Eq. (1):
Among them, represents the surface coverage. reflects the maximum monolayer adsorption. represents the adsorption equilibrium constant. C is the solution concentration. This model precisely controls the arrangement density of molecules at the interface to ensure that each active site is effectively passivated. Surface analysis technology confirms that amino groups form a stable coordination bond network with uncoordinated lead ions in the perovskite lattice, significantly reducing the density of interfacial dangling bonds.
The molecular layer reconstructs the interfacial chemical environment at the atomic scale to form a uniform electronic potential field distribution. Microstructural characterization shows that silane molecules selectively fill ion vacancies at the grain boundaries and block intrinsic migration channels. Dynamic ion tracking experiments confirm that the modified interface exhibits an excellent ion locking effect, effectively inhibiting component segregation driven by the electric field. The interface potential distribution test shows that the charge accumulation phenomenon is significantly alleviated, creating an ideal environment for carrier transport. X-ray photoelectron spectroscopy (XPS) analysis showed that the Pb 4f peak shifted to a lower binding energy direction by about 0.35 eV after APTES treatment, indicating that the amino group formed a coordinate bond with the uncoordinated Pb2+. At the same time, a characteristic peak of 399.2 eV appeared in the N 1s spectrum, corresponding to the N–Pb coordination structure, supporting the proposed ionic-bond-compensation mechanism.
Table 2 systematically summarizes the critical processing parameters and control criteria for interface modification via silane-based self-assembled monolayers (SAMs). In this study, APTES is employed as the precursor molecule. Its concentration is precisely regulated at the millimolar scale to guarantee the optimal surface coverage density of the monolayer. A high-purity isopropanol solvent system (purity ≥ 99.9%) is utilized to provide a favorable environment for the hydrolysis reaction. The immersion time is strictly confined to an 8–12 min window to ensure the completion of molecular self-assembly while preventing multilayer formation. The solution temperature is stably maintained at 25°C ± 0.5°C to preserve the stability of reaction kinetics. Post-treatment annealing is performed at 80°C–100°C to facilitate siloxane condensation and the formation of a robust covalently bonded network. All parameter ranges have been rigorously validated to ensure the construction of a highly ordered monolayer on the perovskite surface without inducing degradation of the underlying perovskite lattice. This established parameter framework offers standardized protocols for realizing reproducible interface engineering, thereby establishing a basis for subsequent device performance optimization. Notably, precise control over temperature and time is essential for fabricating uniform and compact molecular layers, and solvent purity directly governs the quality of the self-assembly process.
| Parameter category | Specific parameter | Control range | Mechanism |
|---|---|---|---|
| Precursor solution | APTES concentration | 1.5-2.5mM | Controlling monolayer coverage density |
| Solvent system | Isopropanol purity | ≥99.9% | Ensuring complete hydrolysis |
| Reaction condition | Immersion time | 8-12min | Guaranteeing complete monolayer formation |
| Temperature control | Solution temperature | 25±0.5°C | Maintaining reaction kinetics stability |
| Post-treatment | Annealing temperature | 80-100°C | Promoting siloxane condensation |
2.1.2. Band regulation and carrier dynamics optimization
While the silane molecular layer passivates defects, its terminal amino group induces the rearrangement of the interface dipole moment, achieving precise control of the band structure. The characteristic shift of the Fermi level position is observed through surface-sensitive energy spectroscopy technology, which significantly improves the energy level matching between the perovskite and the adjacent charge transfer layer. This optimization of the electronic structure significantly reduces the interface non-radiative recombination loss and improves the carrier extraction efficiency.
The carrier dynamics analysis follows the Shockley-Read-Hall recombination theory Eq. (2):
represents the effective carrier lifetime. is the thermal motion velocity. represents the capture cross-section. reflects the trap density. The carrier lifetime is improved by an order of magnitude after interface modification, confirming that the defect-assisted recombination channel is effectively suppressed. Microstructural characterization combined with theoretical calculations reveals that the silane molecules form conformal contact with the perovskite lattice, maintaining the periodicity of the lattice while eliminating the local strain field.
The electron transport performance test shows that the interface contact resistance is significantly reduced, which is attributed to the improvement of band continuity and the reduction of potential energy barrier height. In-situ photoelectric characterization confirms that the modified device maintains a stable charge separation efficiency under continuous operating conditions, and no obvious interface degradation is observed. This robust interface structure provides an ideal platform for achieving efficient and stable carrier transport.
Carrier-dynamics characterization was conducted by time-resolved photoluminescence (TRPL), transient photovoltage (TPV), transient absorption spectroscopy (TAS), and electrochemical impedance spectroscopy (EIS). TRPL was measured using a 400 nm pulsed laser and a single-photon counting system with a temporal resolution of 50 ps under a dry N2 atmosphere. TPV measurements were performed under AM 1.5 white-light bias with a lock-in amplifier at 25 ± 0.5°C. TAS was carried out using a tunable pump-probe system and a spectrometer array with a temporal resolution of 200 fs in a vacuum-sealed chamber. EIS measurements were conducted with a 10 mV AC perturbation over a frequency range of 0.1 Hz to 1 MHz under electromagnetic shielding conditions. All measurements were repeated three times to ensure reproducibility.
2.2. Multi-level interface buffer structure construction
An ultra-thin SnO₂ nanolayer is deposited on the silane-modified layer and used as an electron transport layer and structural buffer layer to regulate the energy band matching and relieve mechanical stress concentration, further improving the interface contact quality and charge extraction efficiency.
2.2.1. Interface energy band engineering of atomic layer deposited SnO₂ nanolayer
Plasma-enhanced atomic layer deposition technology is used to construct SnO₂ nanofilm on the surface of the silane-modified layer under the protection of an inert atmosphere. The alternating cycle of precursor pulses and plasma oxidation steps ensures the self-limiting surface reaction characteristics of film growth. During the deposition process, the reaction chamber temperature and plasma power are precisely controlled to achieve directional epitaxial growth of the SnO₂ lattice. The key to this process is to maintain a moderate oxygen vacancy concentration, which not only ensures the conductivity of the film but also avoids carrier recombination caused by excessive defects.
The regulation of the band structure follows the semiconductor heterojunction theory, and the continuous gradient of the work function is achieved by adjusting the stoichiometric ratio of SnO₂. The electronic structure characteristics of the film satisfy the following relationship Eq. (3):
represents the work function. is the electron affinity. reflects the Fermi level position. Interfacial analysis suggests that the optimized SnO2 layer forms favorable energy-level alignment with the underlying silane-modified layer, thereby facilitating electron transfer across the interface. Synchrotron-based characterization further indicates the presence of a nanoscale elemental interdiffusion region, suggesting the formation of an interfacial transition zone between the two layers. This interfacial structure is consistent with reduced interfacial-state density and improved electronic coupling. Ultraviolet photoelectron spectroscopy (UPS) shows that, after SnO2/silane bilayer modification, the device work function increased from 4.21 eV to 4.48 eV, consistent with the proposed dipole-related band-bending effect at the interface. In addition, grazing-incidence XRD reveals that the full width at half maximum (FWHM) of the perovskite (110) diffraction peak decreased by 18%, suggesting partial relief of lattice strain after interfacial engineering.
Table 3 systematically summarizes the key process parameters and functional objectives of preparing SnO2 nanolayers by plasma-enhanced atomic layer deposition. The precursor used is a combination of SnCl4 and H₂O as a metal source and an oxidant, respectively, and layer-by-layer growth is achieved through alternating pulses. The deposition temperature is controlled in the range of 120°C-150°C, which can maintain sufficient precursor decomposition activity and avoid thermal damage to the perovskite substrate. The SnCl₄ pulse time is limited to 0.1s-0.3s to ensure that the precursor molecules reach saturated adsorption on the substrate surface without producing gas-phase side reactions. The purge time is set to 5s-8s, and the reaction byproducts are completely removed by high-purity nitrogen to ensure the chemical purity of the film and the steepness of the interface. The synergistic effect of various parameters forms a stable self-limiting surface reaction mechanism. Finally, a SnO₂ nano layer with uniform thickness and controllable stoichiometric ratio is obtained. This process window is optimized and verified by orthogonal experiments and shows excellent consistency in repeatability tests, laying the foundation for subsequent band engineering and stress regulation.
| Precursor | SnCl₄/H₂O | Providing Sn source and oxidizer |
|---|---|---|
| Deposition temperature | 120–150°C | Balance reaction kinetics and crystallinity |
| Plasma power | 50–80W | Regulate oxygen vacancy concentration |
| Pulse time (SnCl₄) | 0.1–0.3s | Achieve precursor surface saturation |
| Purge time | 5–8s | Remove unreacted precursors and byproducts |
2.2.2. Mechanical stress coordination and interface stability enhancement
The stress buffering function of the SnO₂ nanolayer originates from its unique microstructure design. Controllable lattice strain is added during the film growth process, and the gradient distribution of the strain state is achieved by precisely controlling the deposition parameters. This strain engineering enables the SnO₂ layer to have excellent stress coordination ability. Its mechanical behavior can be described by the film elasticity theory Eq. (4):
is the actual stress. E is the elastic modulus. represents strain. represents the intrinsic stress. Through microstructural characterization, it is observed that the SnO₂ grains show a special columnar growth morphology, and there is an amorphous phase network in the grain boundary region. This multi-scale structure gives the film excellent mechanical adaptability.
With the action of thermomechanical stress, the SnO₂ buffer layer effectively dissipates stress through mechanisms such as lattice distortion and interface slip. In-situ testing technology confirms that the structure can significantly alleviate the interface stress concentration caused by differences in thermal expansion coefficients. Dynamic mechanical analysis shows that the SnO₂ layer significantly improves the interface binding energy and structural stability of the device under temperature cycling conditions. The unique microstructure of the film also gives the interface excellent resistance to crack propagation, providing a guarantee for the long-term stable operation of the device.
In Figure 2, the horizontal axis represents the interfacial depth (0–10 nm), and the vertical axis represents atomic concentration. The Sn profile shows a gradient distribution in the near-interface region, with a maximum concentration at approximately 2 nm. Its gradual decay is consistent with diffusion-controlled interfacial penetration, suggesting that plasma-assisted deposition promotes shallow incorporation of Sn species into the interfacial region. The Si concentration reaches a maximum of about 50% at around 3 nm, and its distribution extends slightly deeper than that of Sn, which may be associated with partial thermal dissociation and limited diffusion of silane species during deposition. The oxygen concentration decreases in the Sn-rich region, consistent with the growth of the SnO2 layer, and increases again to approximately 65% in the silane-dominated region (>4 nm), indicating that the interfacial process does not cause excessive oxidation of the silane layer.

The observed interdiffusion behavior appears to be governed by both the interfacial chemical potential gradient and the deposition energy. The shallow penetration of Sn may be related to interactions between deposited Sn species and terminal hydroxyl groups of the silane layer, which can promote interfacial anchoring. In contrast, the more limited diffusion of Si is likely constrained by the covalently cross-linked network of the silane molecular layer. These results suggest the formation of an interfacial transition region between the two materials. In addition, oxygen vacancies in the SnO2 lattice may contribute to interfacial electronic regulation through interactions with amino-functional groups in the silane layer, which is consistent with the proposed dipole-related interfacial effect. Such an interlocked interfacial structure may help reduce defect density and alleviate thermal-expansion mismatch between heterogeneous layers. The spatial variation in oxygen concentration also suggests that the interfacial oxidation state remains reasonably controlled, which is beneficial for preserving charge-transport continuity across the interface.
2.3. Low-temperature dynamic thermocompression technology to achieve encapsulation integration
A reversible thermoplastic encapsulant is employed for dynamic thermocompression within the temperature window of 80°C–100°C, enabling molecular-level interdiffusion and structural rearrangement across device encapsulation layers. This strategy substantially enhances the thermomechanical compatibility and barrier performance of the encapsulation system.
2.3.1. Molecular penetration and interpenetration induced by dynamic thermocompression
The thermoplastic encapsulation colloid is rationally designed on the basis of reversible dynamic covalent chemistry, in which furan–maleimide reversible bonding moieties are incorporated into the molecular backbone. During the encapsulation process, the applied thermal gradient triggers the reversible cleavage and reformation of dynamic covalent bonds, endowing the colloid with adaptive flow characteristics in its viscoelastic state. A multi-stage pressure regulation protocol is implemented for the thermocompression procedure. In the initial phase, a moderate pressure is applied to induce viscoelastic deformation of the colloid, thereby filling microscale interfacial voids. Subsequently, a gradient pressure ramp drives the three-dimensional interpenetration of molecular chains, leading to the formation of a cross-scale interpenetrating network structure. The colloid flow behavior follows the constitutive relationship of non-Newtonian fluids Eq. (5):
is shear stress. represents the apparent viscosity. is the shear rate. n is the power law index. Temperature regulation realizes the transition of colloid from solid elastic response to viscous flow plastic behavior by changing the n value. In-situ spectral analysis confirms that the reversible behavior of dynamic covalent bonds promotes the topological reconstruction of molecular chains and forms an interface transition layer chemically bonded to the encapsulation substrate. Molecular dynamics simulation reveals that the terminal functional groups of the colloid form a hydrogen bond-covalent bond hybridization with the hydroxyl groups on the substrate surface, which significantly improves the interface binding energy.
The penetration depth of the colloid is optimized by regulating the interface morphology, and its filling efficiency is controlled by the synergistic effect of the thermocompression temperature field and pressure field. Microstructural characterization shows that the molecular chains formed a gradient entangled network in the interface area, effectively bridging the structural defects from the nanometer to micrometer scale. This molecular-level interlocking mechanism significantly inhibits the tendency of interface delamination and improves the mechanical robustness of the encapsulation system. After thermocompression, the free volume fraction of the interface area is significantly reduced, and the orientation degree and crystallinity of the molecular chains increase synergistically to form a dense barrier.
Table 4 systematically summarizes the key parameters of the dynamic thermocompression process and its regulation mechanism on the molecular penetration behavior. The thermocompression temperature was controlled within 80–100°C, and the pressure was gradually increased from 0.5 to 2 MPa to facilitate interfacial filling and structural rearrangement within the encapsulation stack. Rheological analysis was used to characterize the viscoelastic flow behavior of the encapsulant during processing, while focused ion beam cross-sectional profiling suggested effective filling of interfacial voids from the nanoscale to the microscale. Small-angle X-ray scattering results were consistent with a reduction in free volume and an increase in chain ordering after thermocompression, supporting the formation of a denser barrier structure. Overall, the combined temperature–pressure–shear regulation appears to improve interfacial cohesion and encapsulation compactness. The shear rate range covers 10⁻2s⁻1-101s⁻1. The power law characteristics of non-Newtonian fluids are quantified by the dynamic shear test of the rheometer, revealing the viscoelastic state transition law of the colloid during the thermocompression process. The molecular penetration depth is verified by focused ion beam profile analysis, showing the interface filling effect from the nanometer to the micrometer scale, effectively eliminating the structural defects between the encapsulation layers. Synchrotron radiation small-angle X-ray scattering analysis can show that the free volume of the colloid is reduced after thermocompression, and the orientation degree and crystallinity of the molecular chain are synergistically improved to form a dense barrier network. Through the multi-field coupling of the temperature-pressure-shear field, each parameter synergistically realizes molecular-level interface fusion and structural rearrangement, providing a theoretical basis and process support for the thermomechanical consistency and barrier performance of the encapsulation layer.
| Parameter category | Parameter settings/Observations | Characterization methods | Mechanistic role |
|---|---|---|---|
| Temperature range | 80–100°C | Real-time thermocouple monitoring | Activating dynamic covalent bond exchange |
| Pressure gradient | 0.5MPa→2MPa (gradient ramp) | Feedback-controlled pressure sensors | Driving 3D molecular chain penetration |
| Shear rate range | 10⁻2s⁻1–101 s⁻1 | Rheometer dynamic shear testing | Governing non-Newtonian flow behavior |
| Molecular penetration depth | Nanoscale to microscale filling | Focused ion beam cross-sectional profiling | Eliminating interfacial voids |
| Free volume reduction | Significant decrease (>50%) | Synchrotron small-angle X-ray scattering | Enhancing encapsulation density |
2.3.2. Structural evolution and performance regulation of thermomechanically adapted encapsulation layer
During the dynamic thermocompression process, a periodic mechanical oscillation field is used to induce structural rearrangement of the encapsulation layer through stress-temperature coupling. The colloidal crystallization behavior follows the strain-induced nucleation mechanism. Its kinetic process can be described by the modified Avrami equation Eq. (6):
is the crystallinity. k is the rate constant. m reflects the nucleation growth mode parameter. The mechanical oscillation field changes the spatial distribution of nucleation sites, promoting heterogeneous nucleation and preferential growth of nanocrystals. Wide-angle X-ray diffraction shows that the colloidal molecular chains are arranged along a specific crystal direction to form a nanofiber reinforcement structure that runs through the interface. This anisotropic microstructure gives the encapsulation layer a gradient modulus property. The high crystallinity of the surface area provides rigid protection. The bottom layer retains moderate flexibility to buffer thermal stress.
Coefficient of thermal expansion (CTE) matching of the encapsulation layer is realized via precise modulation of molecular chain segment mobility. The gradient modulus structure effectively dissipates interfacial thermal stress and suppresses the initiation of microcracks induced by thermal cycling. The enhanced barrier performance originates from two synergistic mechanisms: molecular chain densification reduces free-volume pathways, while the oriented arrangement of crystalline domains constructs a tortuous diffusion pathway. Water and oxygen permeability measurements verify that the encapsulation layer exhibits a pronounced inhibitory effect on the adsorption and diffusion of polar molecules. Interfacial chemical-stability analysis suggests that the dynamic-bond network may help accommodate local structural imperfections generated during encapsulation, thereby contributing to sealing integrity during aging.
The thermocompression encapsulation system exhibited improved hygrothermal stability and mechanical robustness under the tested conditions. Taken together, the structural and aging results support the view that interfacial fusion and gradient-structure design contribute to enhanced protection of PSC devices across multiple length scales. Environmental aging tests further indicate that the fluorosilicone coating helps preserve edge integrity under humid-heat conditions, while its dynamic network may hinder crack propagation and slow the formation of permeation pathways. As a result, edge protection contributes to improved device stability during environmental exposure.
2.4. Encapsulation edge protection and hygrothermal stability improvement
On the basis of thermocompression encapsulation, a fluorosilicone copolymer coating is added to the outer layer to improve the hydrophobicity of the interface and the adaptability to thermal expansion and contraction, strengthen the device’s resistance to hot-humid aging, and delay the formation of the interface penetration path.
2.4.1. Gradient hydrophobic interface design of fluorosilicone copolymer coating
The sol-gel method and chemical vapor deposition synergistic process are used to construct a gradient cross-linked fluorosilicone copolymer coating on the surface of the encapsulation layer. The precursor is a mixed system of perfluoroalkylsiloxane and alkoxysilane. The hierarchical assembly of fluorocarbon chains and silicon-oxygen networks is achieved by precisely controlling the kinetics of the hydrolysis-condensation reaction. A lotus-like bionic micro-nanostructure is formed on the coating surface. Its hydrophobic properties follow the Cassie-Baxter wetting model Eq. (7):
is the apparent contact angle. is the intrinsic contact angle. is the solid-liquid contact area fraction. By increasing the density of the fluorocarbon chain by gradient, a dense fluorinated barrier is formed on the surface, and a highly cross-linked silicon-oxygen network is retained in the bottom layer to enhance mechanical strength. Molecular dynamics simulations show that the rigid orientation of the fluorocarbon chain and the flexible movement of the silicon-oxygen chain work together to form a dynamic hydrophobic interface: the fluorine atoms on the surface repel the adsorption of water molecules through the steric effect, and the silicon-oxygen bonds in the bottom layer reconstruct the interface stress through hydrogen bonds.
The chemical bonding characteristics of the coating are achieved through the interface transition layer. During the deposition process, the precursor molecules undergo condensation reactions with the active groups on the surface of the encapsulation layer to form an interpenetrating network connected by covalent bonds. This chemical anchoring effect significantly improves the adhesion of the coating and inhibits interfacial stratification in a humid and hot environment. Microstructural characterization confirms that the coating forms a continuous sealing layer on the edge of the encapsulation, effectively blocking the water and oxygen penetration path.
2.4.2. Dynamic thermomechanical adaptation and diffusion path blocking mechanism
The thermal expansion adaptability of fluorosilicone copolymers is derived from their dynamic covalent bond network design. Reversible disulfide bonds and siloxane units are used to achieve strain adaptation through the thermal response characteristics of the bonding state. The thermomechanical behavior of the material follows the modified Grüneisen relationship Eq. (8):
is the coefficient of thermal expansion. is the Grüneisen parameter. is the heat capacity at constant volume. is the bulk elastic modulus. is the molar volume. The reversible breaking and recombination of dynamic bonds enables the coating to actively regulate internal stress during temperature cycling: at high temperatures, the bonds dissociate and absorb expansion strain, and at low temperatures, the bonds reorganize and release contraction stress. Synchrotron radiation characterization shows that the microstructure of the coating remains highly stable under thermal shock, with no obvious phase separation or crack extension.
The permeation barrier function is achieved through the synergy of multi-scale structures. The high-density fluorocarbon chains on the surface construct a superhydrophobic interface, which significantly increases the activation energy of water molecule diffusion. The bottom gradient cross-linked network restricts molecular migration through free volume regulation. The diffusion kinetics behavior satisfies the modified Fick’s law Eq. (9):
is the concentration of the permeate. is the diffusion coefficient. is the adsorption rate constant. The low-polarity surface of the fluorosilicone coating inhibits the adsorption of water molecules, and the tortuous diffusion path prolongs the penetration time. Environmental aging tests show that the coating maintains stable interface integrity under humid and hot conditions, and its dynamic bond network can repair microcracks and block the expansion of the penetration channel. The enhanced protection of the encapsulation edge enables the device to maintain excellent performance stability in extreme environments.
3. Results and Discussion
3.1. Modification effect, transfer characteristics and interface defect density determination
Figure 3 shows the scanning electron microscope (SEM) characterization results (300nm) of the surface morphology of perovskite films with different interface modification strategies. The surface of the unmodified perovskite film shows significant non-uniformity, with a wide distribution of grain sizes, irregular shapes, noticeable cracks, and pinhole defects at the grain boundaries, and low surface coverage, indicating that the rapid crystallization process under unmodified conditions leads to defect enrichment. The surface morphology of the film modified with APTES is significantly improved. The grain size is uniform, and it presents a tightly arranged hexagonal structure. There are no visible cracks or pinholes at the grain boundaries, and the surface coverage is close to complete. The silane molecular layer effectively passivates the uncoordinated defects through chemical bonding to form a continuously covered passivation network. A two-dimensional/three-dimensional mixed-phase structure is observed on the surface of the PEAI-modified film. The surface of the bottom three-dimensional perovskite grains is covered with an ultra-thin two-dimensional layer. However, the two-dimensional phase is unevenly covered in local areas, and sporadic pinholes remain, indicating that some defects are not completely shielded due to kinetic limitations during solution spin coating. The NiOx-modified film shows that the perovskite grains grow epitaxially along the substrate, but there is lattice distortion, and parallel microcracks are visible at the interface. The comparison of the four groups of morphologies intuitively reflects the differences in the effects of different modification strategies on the defect suppression and interface stability of perovskite films.

During the evaluation process, the space-charge-limited current (SCLC) method is used. The current-voltage characteristics of the device are tested under dark conditions. The voltage range of 0-5V is scanned by the Keithley 4200 semiconductor analysis system, and the transition inflection point between the trap filling region (Child’s law region) and the ohmic conduction region is recorded. Ten devices are prepared for each of the four groups of samples: modified with APTES, unmodified, modified with PEAI, and modified with NiOx for repeated testing. In the data processing stage, the trap filling voltage is determined by the change in the slope of the current density and voltage curve under double logarithmic coordinates. The trap density is calculated using Eq. (10):
is the relative permittivity. is the vacuum permittivity. e is the elementary charge. is the thickness of the active layer. The test environment is controlled at 25°C and in a nitrogen atmosphere to avoid environmental interference.
Figure 4 presents the charge-transfer characteristics and interfacial defect density of PSCs with different interface modification strategies. The SCLC curves provide insight into how interfacial engineering influences carrier-transport behavior. The unmodified device exhibits superlinear current growth in the low-voltage region, indicating that carrier transport is strongly affected by trap-assisted capture and release processes associated with interfacial defect states. As the applied voltage increases, the current gradually approaches the cubic dependence predicted by the SCLC model, reflecting the transition to space-charge-limited conduction. In contrast, the APTES-modified device shows a current response closer to ideal SCLC behavior over the entire voltage range, suggesting that the silane self-assembled monolayer effectively suppresses trap-assisted scattering and recombination through interfacial chemical passivation involving amino–Pb2⁺ coordination. The PEAI- and NiOx-modified devices exhibit intermediate behavior, indicating partial improvement in charge transport; this can be associated with the dielectric screening effect of the 2D perovskite capping layer in the PEAI-treated device and the altered interfacial energetics introduced by the NiOx layer.

The trap-density statistics further clarify the influence of interface modification on defect suppression. In Figure 4(b), the black, purple, magenta, and green curves represent the unmodified, APTES-treated, PEAI-treated, and NiOₓ-modified devices, respectively. These coloured curves show the trap-state-density values obtained during the repeated test sequence for each device group and illustrate the corresponding test-to-test variation; they do not represent confidence intervals or uncertainty bands. The APTES-modified group exhibits the lowest trap density, reaching 2.1 × 101⁶ cm⁻3, with relatively small dispersion, indicating improved interfacial uniformity after molecular self-assembly treatment. By comparison, the PEAI-modified group shows larger fluctuations, which may be related to the spatial inhomogeneity of the 2D/3D perovskite phase distribution during solution processing and the resulting local variation in carrier-transport pathways. The NiOx-modified group also displays noticeable defect-density variation, which can be associated with interfacial heterogeneity and possible lattice mismatch effects that influence the quality of perovskite growth on the oxide surface. From the perspective of carrier dynamics, the APTES-modified interface appears to reduce the barrier for electron extraction by improving interfacial band alignment, whereas PEAI and NiOx, although beneficial to energy-level arrangement in certain respects, remain limited by residual interfacial defects.
Taken together, these two sets of data highlight the distinction between chemically bonded interfacial regulation and more physically dominated passivation modes. The silane-derived interfacial network provides improved structural and electronic stability, helping to suppress defect-state regeneration associated with ion migration under operating stress. By contrast, interfaces relying more strongly on physical contact or less stable interfacial interactions are more susceptible to degradation under hygrothermal conditions, which can lead to a gradual loss of passivation effectiveness over time. This difference is further reflected in the subsequent durability results, underscoring the importance of molecular-level interfacial engineering for maintaining PSC performance under complex operating conditions.
3.2. Carrier lifetime and complex dynamic analysis
Based on TRPL and TPV decay tests, the changes in the carrier lifetime of the device under different interface treatments are quantified, and the interface’s ability to suppress the recombination process is evaluated.
Figure 5 shows the TRPL carrier-lifetime distributions under four different interface-treatment conditions, where the horizontal axis represents carrier lifetime (ns), and the vertical axis represents normalized frequency. The unmodified sample exhibits a broad lifetime distribution, which can be attributed to the presence of unpassivated ionic defects on the perovskite surface and randomly distributed deep-level trap states at grain boundaries. These defects act as non-radiative recombination centers and significantly shorten the intrinsic carrier lifetime. The PEAI-modified group shows a shift toward longer lifetimes, suggesting partial passivation of surface defects through the interaction between phenylethylammonium species and surface lead vacancies. However, the pronounced tail in the distribution indicates that this passivation route may provide limited relief of interfacial strain and incomplete spatial uniformity. Although NiOx modification can improve charge extraction by optimizing energy-level alignment, interfacial stress associated with lattice mismatch may introduce additional recombination pathways. In contrast, the APTES-modified group exhibits the narrowest and most right-shifted lifetime distribution, indicating more effective suppression of non-radiative recombination. This behavior is consistent with the formation of a chemically bonded interfacial network, in which amino–Pb2⁺ coordination contributes to defect passivation, while the Si–O–Si framework may help accommodate local thermomechanical mismatch. The improved symmetry and narrowing of the APTES lifetime distribution further suggest enhanced spatial uniformity of interfacial passivation, which is beneficial for maintaining device consistency over larger active areas. Overall, the evolution of the TRPL distributions highlights the advantages of chemically bonded interface engineering for improving carrier dynamics in PSC devices.

Figure 6 presents the TPV lifetime distributions for the same four interface-treatment conditions. Compared with TRPL, TPV is more sensitive to interfacial recombination processes, and shortened TPV lifetimes generally reflect increased non-radiative carrier loss at the interface. The unmodified sample shows a distinctly left-shifted and broadened distribution, indicating severe interfacial recombination associated with unpassivated dangling bonds and metastable defects induced by ion migration at the perovskite/electrode interface. The PEAI-modified group displays improved lifetimes, indicating partial suppression of surface recombination; however, because the interaction between the organic capping species and the perovskite surface is relatively weak, structural relaxation under prolonged electrical stress may expose incompletely passivated regions. For the NiOx-modified device, improved band alignment is accompanied by increased interfacial heterogeneity arising from lattice mismatch, which may create additional recombination channels and reduce the overall passivation effectiveness. By contrast, the APTES-modified device shows the longest and most concentrated TPV lifetime distribution, suggesting more efficient suppression of interfacial recombination while maintaining favorable charge-transport characteristics. This improvement is consistent with a more stable chemically bonded interface, which helps stabilize interfacial ions and reduce local recombination variability. In addition, the narrowing of the TPV distribution implies improved spatial uniformity of the recombination rate across the interface. Compared with the other modification strategies, APTES treatment appears to provide a more balanced combination of interfacial passivation and carrier-extraction efficiency, supporting its effectiveness for improving PSC stability and performance.

3.3. Accelerated thermal and damp-heat stability evaluation
To evaluate the environmental durability of the encapsulated PSC devices, accelerated aging tests were conducted using an industrial-grade temperature–humidity chamber. Four encapsulation strategies were compared: the dynamic thermocompression encapsulation proposed in this work, conventional glass encapsulation, epoxy-resin encapsulation, and hot-melt adhesive encapsulation. The aging tests were carried out under two conditions: (i) thermal aging at 85°C under dry conditions (<5% RH) and (ii) damp-heat aging at 85°C/85% RH, with a total duration of 1000 h. Five parallel samples were tested for each group. The devices were removed at regular intervals for photovoltaic characterization, and the current–voltage curves were measured under standard AM 1.5G illumination using a Keithley 2400 source meter and a Class AAA solar simulator (Oriel Sol3A, 1000 W m⁻2). These conditions were selected to enable a more meaningful comparison with widely adopted accelerated stability benchmarks for photovoltaic devices.
Figure 7 shows the evolution of power-conversion efficiency (PCE) of the four encapsulation systems during 1000 h of thermal aging at 85°C. The results reveal pronounced differences in thermomechanical stability among the encapsulation strategies. The glass-encapsulated devices exhibit progressive efficiency decay, which can be attributed to interfacial stress accumulation caused by thermal-expansion mismatch. The epoxy-resin group also shows continuous degradation, likely associated with curing-induced internal stress and subsequent thermal embrittlement. The hot-melt adhesive system undergoes the fastest performance loss, consistent with insufficient thermal resistance and structural instability at elevated temperatures. In contrast, the dynamically thermo-compressed devices maintain approximately 96% of their initial PCE after 1000 h, indicating substantially improved resistance to thermal degradation. This enhanced stability is attributed to the adaptive stress-dissipation capability of the dynamic encapsulation network and the improved interfacial integrity enabled by the coordinated encapsulation design.

Figure 8 presents the damp-heat stability results under the more stringent condition of 85°C/85% RH for 1000 h. Compared with the thermal-aging test, all conventional encapsulation systems exhibit more severe efficiency loss, reflecting the combined impact of moisture ingress, oxygen penetration, and hygrothermal interfacial degradation. The glass-encapsulated and epoxy-resin devices show marked deterioration during prolonged exposure, while the hot-melt adhesive group undergoes the most rapid degradation, indicating limited barrier capability against water and oxygen under humid conditions. By contrast, the dynamically thermocompressed PSCs retain approximately 92.5% of their initial power conversion efficiency (PCE) after 1000 hours of damp-heat aging. This result suggests that the combination of low-temperature dynamic encapsulation and fluorosilicone edge protection effectively suppresses moisture/oxygen ingress and mitigates thermomechanical damage at the encapsulation interface. Although this 1000 h test substantially strengthens the durability assessment relative to shorter preliminary evaluations, it should still be regarded as an accelerated device-level stability test rather than a full certification of long-term field operation. Nevertheless, the results provide strong evidence that the proposed interface–encapsulation strategy significantly enhances PSC durability under harsh environmental stress.

3.4. Water and oxygen permeability determination
During the evaluation process, a 50nm thick calcium film (deposited on a glass substrate) is integrated inside the encapsulated sample. A UV-visible spectrophotometer (PerkinElmer Lambda 950) is used to monitor the transmittance changes of the calcium film at a characteristic wavelength of 658nm in real time. The test system integrated an in-situ humidity sensor (Honeywell HIH-4000) and an oxygen concentration monitor (Systech EC910). Transmittance, humidity, and oxygen concentration data are collected every 5 min. The oxidation reaction rate of the calcium film is linearly related to the transmittance attenuation, and the water vapor transmission rate (WVTR) and oxygen transmission rate (OTR) are measured. The data are processed by Savitzky-Golay filtering to eliminate noise interference.
Figure 9 illustrates the protective performance of different encapsulation strategies for the environmental stability of perovskite devices, as evaluated by WVTR and oxygen transmission rate (OTR) measurements. In both panels, the black, purple, magenta, and green curves represent glass encapsulation, epoxy encapsulation, hot-melt adhesive encapsulation, and dynamic thermocompression encapsulation, respectively. Specifically, the coloured curves in Figure 9(b) show the time-dependent OTR responses of the four encapsulation systems and do not represent confidence intervals or uncertainty bands. The logarithmic comparison of WVTR and OTR indicates that conventional encapsulation approaches are constrained by intrinsic material limitations and process-related defects. In the glass-encapsulation system, the relatively high WVTR is mainly attributed to thermal-expansion mismatch between the metal solder and the glass substrate, which promotes interfacial microcrack formation under cyclic stress and provides rapid diffusion pathways for water vapor. For epoxy-resin encapsulation, the polar polymer chains absorb ambient moisture through hydrogen bonding, and the resulting swelling further enlarges free-volume regions, thereby facilitating water penetration. In the hot-melt adhesive system, the porous structure directly forms continuous permeation pathways, and the concurrent deterioration in WVTR and OTR reflects the inherent limitation of this physically sealed structure in balancing barrier performance with mechanical flexibility. By contrast, the substantially lower green curves in Figures 9(a) and 9(b) demonstrate that dynamic thermocompression encapsulation provides the strongest resistance to both water-vapor and oxygen permeation among the four encapsulation strategies. In particular, the OTR of the dynamically thermocompressed system approaches the lower detection limit of the instrument, supporting the formation of a compact and highly effective environmental barrier.

By contrast, the dynamic thermocompression encapsulation exhibits markedly improved barrier properties, which can be attributed to the chemically and mechanically coordinated design of the functional coating. The gradient cross-linked fluorosilicone network forms a dense fluorocarbon-rich surface barrier that suppresses water adsorption, while the underlying dynamic siloxane network helps relieve thermally induced microcracks through bond reorganization and maintains structural compactness. The very low OTR further supports the dual-barrier function of this architecture: the ordered fluorocarbon segments hinder oxygen penetration, whereas the entangled siloxane network increases the diffusion resistance by regulating free volume. This rigid-flexible integrated structure enables the encapsulation layer to maintain continuity under deformation, thereby reducing permeation channels caused by bending or thermal expansion/contraction. Overall, the permeability of the dynamic thermocompression encapsulation approaches the lower detection limit of the instrument, indicating substantially enhanced resistance to water and oxygen ingress. In essence, this superior barrier performance arises from reducing diffusion pathways from micron-scale interfacial defects to sub-nanometer molecular-scale gaps, thereby slowing the degradation kinetics of the device.
4. Conclusions
This study presents a coordinated materials-and-device engineering strategy to enhance the stability of PSCs for future energy-storage-oriented applications. By integrating silane self-assembly interfacial passivation, an SnO₂ buffer layer, low-temperature dynamic thermocompression encapsulation, and fluorosilicone edge protection, the proposed framework simultaneously addresses two coupled degradation pathways that are highly relevant to PSC durability, namely interfacial ionic/defect-related degradation and encapsulation-related thermomechanical/environmental failure. XPS, UPS, and grazing-incidence XRD analyses support the proposed interfacial coordination, work-function regulation, and strain-relief effects associated with the interface-engineering design. Device-level characterization further shows reduced trap-state density, improved charge-transport behavior, and enhanced encapsulation integrity.
Under accelerated damp-heat aging at 85°C/85% RH, the optimized PSC devices retain over 92.5% of their initial power conversion efficiency after 1000 h, indicating substantially improved environmental durability compared with the control strategies. In this sense, the contribution of the present work lies not in claiming each functional component as entirely unprecedented, but in establishing a synergistic interface–buffer–encapsulation–edge-protection framework for PSC stability enhancement through coordinated materials and device engineering. These results provide a practical device-level basis for future exploration of PSC deployment in energy-storage-related application scenarios.
Although the 1000 h damp-heat results substantially strengthen the durability assessment, the present study remains a laboratory-scale materials and device investigation rather than a full system-level validation. Future work should therefore include coupled PSC–storage operation, charge/discharge cycling, outdoor testing, and application-level performance evaluation to further assess long-term field stability and practical integration potential.
Acknowledgment
This work was financially supported by the China Scholarship Council (No.202508220056).
CRediT authorship contribution statement
Jiaxuan Ren: Conceptualization, Methodology, Investigation, Data curation, Formal analysis, Writing – original draft, Writing – review & editing, Supervision, Project administration. Tongyu Zhu: Methodology, Investigation, Data curation, Formal analysis, Visualization, Writing – review & editing.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declaration of generative AI and AI-assisted technologies in the writing process
The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.
References
- Stability of perovskite solar cells: Issues and prospects. RSC Advances. 2023;13:1787-1810. https://doi.org/10.1039/d2ra05903g
- [Google Scholar]
- Toward efficient perovskite solar cells: Progress, strategies, and perspectives. ACS Energy Letters. 2022;7:2084-2091. https://doi.org/10.1021/acsenergylett.2c00592
- [Google Scholar]
- Efficient perovskite solar cells via improved carrier management. Nature. 2021;590:587-593. https://doi.org/10.1038/s41586-021-03285-w
- [Google Scholar]
- Buried interface molecular hybrid for inverted perovskite solar cells. Nature. 2024;632:536-542. https://doi.org/10.1038/s41586-024-07723-3
- [Google Scholar]
- The high open-circuit voltage of perovskite solar cells: A review. Energy & Environmental Science. 2022;15:3171-3222. https://doi.org/10.1039/d2ee00663d
- [Google Scholar]
- A review on perovskite solar cells (PSCs), materials and applications. Journal of Materiomics. 2021;7:940-956. https://doi.org/10.1016/j.jmat.2021.04.002
- [Google Scholar]
- Highly efficient p-i-n perovskite solar cells that endure temperature variations. Science (New York, N.Y.). 2023;379:399-403. https://doi.org/10.1126/science.add7331
- [Google Scholar]
- Homogenizing out-of-plane cation composition in perovskite solar cells. Nature. 2023;624:557-563. https://doi.org/10.1038/s41586-023-06784-0
- [Google Scholar]
- Double-side 2D/3D heterojunctions for inverted perovskite solar cells. Nature. 2024;628:93-98. https://doi.org/10.1038/s41586-024-07189-3
- [Google Scholar]
- Rapid advances enabling high-performance inverted perovskite solar cells. Nature Reviews Materials. 2024;9:399-419. https://doi.org/10.1038/s41578-024-00678-x
- [Google Scholar]
- Surface reaction for efficient and stable inverted perovskite solar cells. Nature. 2022;611:278-283. https://doi.org/10.1038/s41586-022-05268-x
- [Google Scholar]
- Molecular engineering of contact interfaces for high-performance perovskite solar cells. Nature Reviews Materials. 2023;8:89-108. https://doi.org/10.1038/s41578-022-00503-3
- [Google Scholar]
- Inverted perovskite solar cells using dimethylacridine-based dopants. Nature. 2023;620:545-551. https://doi.org/10.1038/s41586-023-06207-0
- [Google Scholar]
- Molecular cation and low-dimensional perovskite surface passivation in perovskite solar cells[J] Nature Energy. 2024;9:779-792. https://doi.org/10.1038/s41560-024-01529-3
- [Google Scholar]
- Aqueous synthesis of perovskite precursors for highly efficient perovskite solar cells. Science (New York, N.Y.). 2024;383:524-531. https://doi.org/10.1126/science.adj7081
- [Google Scholar]
- Tailoring passivators for highly efficient and stable perovskite solar cells. Nature Reviews. Chemistry. 2023;7:632-652. https://doi.org/10.1038/s41570-023-00510-0
- [Google Scholar]
- Amidino-based Dion-Jacobson 2D perovskite for efficient and stable 2D/3D heterostructure perovskite solar cells. Joule. 2023;7:574-586. https://doi.org/10.1016/j.joule.2023.02.003
- [Google Scholar]
- Origins and influences of metallic lead in perovskite solar cells. Joule. 2022;6:816-833. https://doi.org/10.1016/j.joule.2022.03.005
- [Google Scholar]
- Improved charge extraction in inverted perovskite solar cells with dual-site-binding ligands. Science (New York, N.Y.). 2024;384:189-193. https://doi.org/10.1126/science.adm9474
- [Google Scholar]
- Dimensional regulation from 1D/3D to 2D/3D of perovskite interfaces for stable inverted perovskite solar cells. Journal of the American Chemical Society. 2024;146:7555-7564. https://doi.org/10.1021/jacs.3c13576
- [Google Scholar]
- 2D Perovsktie substrate-assisted CsPbI3 film growth for high-efficiency solar cells. ACS Applied Materials & Interfaces. 2022;14:7417-7427. https://doi.org/10.1021/acsami.1c20968
- [Google Scholar]
- Direct surface passivation of perovskite film by 4-fluorophenethylammonium iodide toward stable and efficient perovskite solar cells. ACS Applied Materials & Interfaces. 2021;13:2558-2565. https://doi.org/10.1021/acsami.0c17773
- [Google Scholar]
- Addressing the voltage induced instability problem of perovskite semiconductor detectors. ACS Energy Letters. 2022;7:3871-3879. https://doi.org/10.1021/acsenergylett.2c02054
- [Google Scholar]
- Surface redox engineering of vacuum-deposited NiOx for top-performance perovskite solar cells and modules. Joule. 2022;6:1931-1943. https://doi.org/10.1016/j.joule.2022.06.026
- [Google Scholar]
- Effect of UVO Treatment on optical and electrical properties of NiOx thin film and perovskite solar cells[J] Current Photovoltaic Research. 2024;12:1-5. https://doi.org/10.21218/CPR.2024.12.1.001
- [Google Scholar]
- Review of technology specific degradation in crystalline silicon, cadmium telluride, copper indium gallium selenide, dye sensitised, organic and perovskite solar cells in photovoltaic modules: Understanding how reliability improvements in mature technologies can enhance emerging technologies. Progress in Photovoltaics: Research and Applications. 2022;30:1365-1392. https://doi.org/10.1002/pip.3577
- [Google Scholar]
- Multifunctional acrylic polymers with enhanced adhesive property serving as excellent edge encapsulant for stable optoelectronic devices. ACS Applied Materials & Interfaces. 2024;16:5138-5148. https://doi.org/10.1021/acsami.3c16598
- [Google Scholar]
- An effective encapsulation for perovskite solar cells based on building-integrated photovoltaics. Journal of Materials Chemistry C. 2022;10:8972-8978. https://doi.org/10.1039/d2tc01696f
- [Google Scholar]
- Silane doping for efficient flexible perovskite solar cells with improved defect passivation and device stability. ACS Applied Materials & Interfaces. 2024;16:23668-23676. https://doi.org/10.1021/acsami.4c00887
- [Google Scholar]
- Bandgap-universal passivation enables stable perovskite solar cells with low photovoltage loss. Science (New York, N.Y.). 2024;384:767-775. https://doi.org/10.1126/science.ado2302
- [Google Scholar]
- Engineering of buried interfaces in perovskites: Advancing sustainable photovoltaics. Nano Convergence. 2024;11:57. https://doi.org/10.1186/s40580-024-00464-z
- [Google Scholar]
- Toward enhancing the surface adhesion of fluoropolymer-based coating materials. Polymer Reviews. 2024;64:980-1030. https://doi.org/10.1080/15583724.2024.2352360
- [Google Scholar]
- Membrane-free redox flow cell based on thermally regenerative electrochemical cycle for concurrent electricity storage, cooling and waste heat harnessing of perovskite solar cells. Journal of Power Sources. 2022;548:232081. https://doi.org/10.1016/j.jpowsour.2022.232081
- [Google Scholar]
- Hard carbon anode materials for hybrid sodium-ion/metal batteries with high energy density. Journal of Power Sources. 2024;624:235547. https://doi.org/10.1016/j.jpowsour.2024.235547
- [Google Scholar]
- MOFs based on the application and challenges of perovskite solar cells. iScience. 2021;24:103069. https://doi.org/10.1016/j.isci.2021.103069
- [Google Scholar]
