# Fileset

[Solar RRL - 2025 - Ishikawa - Outdoor Performance and Thermally Accelerated Degradation of Inverted Perovskite Solar Cells.pdf](https://mdr.nims.go.jp/filesets/eb84ea48-2ba5-49a1-860d-534982cb9668/download)

## Creator

Ryousuke Ishikawa, Hayato Okawa, [Masatoshi Yanagida](https://orcid.org/0000-0002-8065-7875), [Yasuhiro Shirai](https://orcid.org/0000-0003-2164-5468), Makoto Konagai

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Outdoor Performance and Thermally Accelerated Degradation of Inverted Perovskite Solar Cells](https://mdr.nims.go.jp/datasets/5aaa4f24-3348-4f61-baac-8a761d259bd1)

## Fulltext

Outdoor Performance and Thermally Accelerated Degradation of Inverted Perovskite Solar CellsOutdoor Performance and Thermally AcceleratedDegradation of Inverted Perovskite Solar CellsRyousuke Ishikawa1,2 | Hayato Okawa1 | Masatoshi Yanagida3 | Yasuhiro Shirai3 | Makoto Konagai21Department of Electrical, Electronic and Communication Engineering, Tokyo City University, Setagaya-ku, Japan | 2Advanced Research Laboratories,Tokyo City University, Setagaya-ku, Japan | 3Photovoltaic Materials Group, Center for GREEN Research on Energy and Environmental Materials, NationalInstitute for Materials Science (NIMS), Tsukuba, JapanCorrespondence: Makoto Konagai (mkonagai@tcu.ac.jp)Received: 6 July 2025 | Revised: 16 September 2025 | Accepted: 28 September 2025Funding: Priority Research Project of Tokyo City UniversityKeywords: outdoor performance | perovskite solar cells | thermal phase transitionABSTRACTIn this article, the long-term outdoor performance of inverted perovskite solar cells (PSCs) with glass–glass encapsulation isinvestigated over a 3-year period. Despite achieving initial power conversion efficiencies of>20%, the devices show significantdegradation during summer months, particularly when surface temperatures exceed 50°C, with their internal cell temperaturesincreasing to ~90°C. Optical and scanning electron microscopy imaging reveal the formation of a high-resistance phase within theperovskite layer, particularly near the NiOx interface, as the major degradation pathway. Indoor accelerated tests, conductedunder heat and light conditions, replicate these effects, indicating the occurrence of thermally induced phase transitions.The measurement methodology also influences the degradation; devices measured with maximum power point tracking showslightly less deterioration than do those assessed via current–voltage curve tracing. Additionally, UV-cut filters show minimalbenefits, likely owing to the inherent UV blocking feature of NiOx. These findings underscore the critical role of thermal man-agement and operational conditions in ensuring the stability of PSCs. Further improving the material design, encapsulation, andmeasurement protocols is essential for enhancing the outdoor reliability and commercial viability of PSCs.1 | IntroductionPerovskite solar cells (PSCs) have garnered considerable atten-tion in recent years because of their rapidly advancing powerconversion efficiencies, facile fabrication processes, and potentialfor low-cost lightweight photovoltaic applications. Laboratory-scale devices have achieved efficiencies exceeding 25%, whichare comparable to those of conventional silicon-based solar cells[1–3]. Among the various device architectures, the inverted (p–i–n)structure has gained interest owing to its compatibility with flexi-ble substrates and reduced hysteresis behavior [4–6].Despite their promising attributes, the practical deployment ofPSCs in outdoor environments remains a significant challenge.Notably, the long-term stability of perovskite materials underreal-world operating conditions, including variable tempera-tures, humidities, solar irradiation, and electrical loads, hasnot yet been systematically and comprehensively examined.Although numerous studies on PSC degradation under con-trolled indoor conditions have been reported to date [7–11],its long-term behavior in actual outdoor environments remainsunderexplored [12–15].Outdoor degradation of PSCs is primarily driven by temperaturefluctuations, UV exposure, and electrochemical stress, whichlead to failure modes such as ion migration, film cracking,and interface degradation [11, 16]. High operating tempera-tures, often exceeding 80°C during peak sunlight hours,This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the originalwork is properly cited, the use is non-commercial and no modifications or adaptations are made.© 2025 The Author(s). Solar RRL published by Wiley-VCH GmbH.Solar RRL, 2025; 0:e202500524 1 of 7https://doi.org/10.1002/solr.202500524Solar RRLRESEARCH ARTICLEhttps://orcid.org/0000-0002-3115-0058mailto:mkonagai@tcu.ac.jphttp://creativecommons.org/licenses/by-nc-nd/4.0/https://doi.org/10.1002/solr.202500524https://doi.org/10.1002/solr.202500524http://crossmark.crossref.org/dialog/?doi=10.1002%2Fsolr.202500524&domain=pdf&date_stamp=2025-10-09reportedly accelerate the degradation of the perovskite layer,resulting in significant efficiency losses [12]. Previous studiesdemonstrate that high-temperature stress can induce the for-mation of low-conductivity phases within perovskite materials,thereby contributing to their long-term performance deteriora-tion [17–19].In this study, we investigated the outdoor stability of encapsu-lated PSCs with an inverted architecture: glass/indium–tin oxide(ITO)/NiOx/perovskite/C60/bathocuproine (BCP)/Ag. We per-formed multiyear outdoor monitoring under different installa-tion configurations and conducted a comparative analysisbased on accelerated indoor degradation tests to identify thekey degradation pathways. We specifically focused on the impactof high temperatures, UV exposure, and measurement method-ologies (current–voltage (I–V ) curve tracing vs. maximum powerpoint tracking (MPPT)) on the long-term performance of PSCdevices.Our findings demonstrated that high cell temperatures, whichfrequently exceeded 80°C during the summer within the studyperiod, play a critical role in inducing resistive phases withinthe perovskite layer and thus lead to substantial efficiency losses.Comparable degradation features were reproduced under con-trolled indoor heating and illumination conditions, suggestingthat thermally driven phase transitions, potentially coupled withcharge accumulation, were central to the observed degradationmechanisms. The investigated correlation between the outdoorand indoor degradation behaviors provides insights into PSC reli-ability under realistic operating conditions. These findings high-light the necessity for developing effective thermal managementand measurement protocols to achieve long-term stability inperovskite photovoltaics.2 | Results and Discussion2.1 | Long-Term Outdoor Degradation BehaviorFigure 1 illustrates the degradation trend of the devices over 3years. The initial efficiencies ranged from 20% to 23% (Figure S1and Table S1), and all the devices exhibited notable degradationduring the summer. New samples, measured in April 2023using a four-terminal configuration, showed initial efficienciesof>20% and fill factors of ~0.8. A significant performance losswas observed in July 2023 and August 2023. The samples sub-jected to two-axis tracking exhibited an earlier onset of degrada-tion than did those with a fixed-tilt configuration.To assess the impact of UV light, the devices were exposed tooutdoor conditions with and without UV-cut filters (cut-off< 400nm). The quantum efficiency (QE) spectrum of the filter is pre-sented in Figures 2a, and 2b shows the daily efficiency degrada-tion as a function of accumulated solar exposure. The negligibleeffect of the UV filter may be attributed to the UV transmittanceof NiOx. The transmittance spectra shown in Figure S2 indicatethat ~20 nm-thick NiOx blocks approximately 10% of UV light(wavelength<400 nm) [20].2.2 | Comparison between I–V Curve Tracing andMPPTBeginning in April 2024, MPPT measurements were conductedalongside I–V curve tracing owing to concerns that I–V curvetracing may contribute to degradation. Figure 3 compares theMPPT measurement and I–V curve tracing results. The MPPTmeasurements continuously tracked the maximum power point,FIGURE 1 | Long-term outdoor degradation trend.2 of 7 Solar RRL, 2025 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewhereas, the I–V curves were measured only three times daily at11:00, 12:00, and 13:00 JST. The daily average efficiency (definedas the total amount of electricity generated per day divided by theaccumulated solar radiation on the same day) data were plottedonly for days with solar irradiance greater than 5 kWh/m2.Although the number of samples was limited, the MPPT resultedin slightly less degradation than did the I–V curve tracingmethod. The I–Vmeasurements were performed in an open state;therefore, the behavior of the carriers inside the cell was differentfrom that revealed by the MPPT method [21]. Moreover, the celltemperatures were possibly different.2.3 | Temperature DependanceFigure 4 compares the temperature and conversion efficiencymeasured at 12:00 using the I–V curve tracing method for thesouth-facing tilt angle of 35°. The temperature trend showsthe surface temperature at 12:00 and the instantaneous maxi-mum temperature of the day (Figure 4a). The conversion effi-ciency data are plotted for an irradiation intensity ofapproximately 100 mW/cm2 at 12:00 h (Figure 4b). For PSCs,a large fill factor is observed under low irradiation intensities,even after degradation. A comparison of the data collected undermeteorological conditions, with irradiation intensities above acertain threshold, reveals that both the methods yield similarresults. The examined trends of Rsh and Rs, shown in FigureS3, indicate that in late July, Rs increases, while Rsh decreasessharply, concurrent with the increase in temperatures.The data indicated that conversion efficiency and fill factors grad-ually degraded fromApril 2024, with rapid degradation occurringwhen the surface temperature exceeded 50°C in July and August2024. Furthermore, the initially minimal hysteresis graduallyincreased. When no temperature correction was applied to thesolar cells, the conversion efficiency decreased from its initialvalue of 20% to 10%–15% by the end of August 2024.Organic-halide perovskite materials are highly susceptible tomoisture, humidity, temperature, and solar radiation, which sig-nificantly affect the degradation process. Figure S4 compares thehumidity data for the same period shown in Figure 4. The humid-ity data are plotted as the average and maximum daily humiditymeasured using a hygrometer installed at the same location asthat of the solar cell. Because temperature and humidity are cor-related to Tokyo’s climate, a strong correlation is expectedbetween performance degradation and humidity. However, nodistinct change in the humidity was observed during the periodsof accelerated degradation. Furthermore, the encapsulated cellsused in this study maintained an initial efficiency of 88.48% after1000 h of continuous irradiation at 1 sun in an MPPT environ-ment with T= 60°C and RH= 30%–35% (as shown in by Khadkaet al. [6]). These factors demonstrate the negligible effect of mois-ture penetration on the rapid degradation modes of these devices.2.4 | Estimation of Cell TemperatureTo estimate the internal cell temperature, the open-circuitvoltage (Voc) was analyzed as a function of the surface temper-ature of the devices (Figure S5). The thermal coefficient wasΔVoc = − 1.8mV°C , which was consistent with reported values[12, 22].In this study, the sample was sealed with glass, and the wholesystem was solidified with resin and adhesive; therefore, the finalFIGURE 2 | (a) QE spectrum and (b) daily efficiency degradation of PSC with the UV filter.FIGURE 3 | Daily efficiency degradation due to MPPT and I–V curvetracing measurements.3 of 7 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensestructure was extremely prone to heat buildup. In other words, alarge discrepancy was expected between the measured surfacetemperature and actual ‘cell temperature’ at the junction.Therefore, in outdoor measurements, we estimated the ‘cell tem-perature’ at high temperatures from the Isc–Voc measurementresults obtained on a sunny day [23]. Figure 5 shows that theVOC values for the 35° sample and for the sample with two-axistracking shift by 60 and 100mV, respectively, which correspondto temperature increments of 33 and 56°C (determined by con-verting the temperature characteristic results into temperaturevalues), respectively. This result suggests that the cell tempera-ture increases to ~90°C during the hot summer months.2.5 | Microscopic and Structural ChangesOptical microscopy (OM) images revealed a localized contrastchange for the sample deteriorated outdoors (Figure 6b). Theratio of spots with different contrasts in the OM images wasquantified; more areas with a high contrast in the image yieldeda higher ‘Mean.Value.’ For example, the Mean.Value changedfrom 17.84 to 29.62 before and after outdoor exposure(Figure 6a and 6b), respectively. In addition, scanning electronmicroscopy (SEM) visualization of the cross-section of this sam-ple revealed the presence of a highly resistive phase in the perov-skite layer (Figure 7). When the same field of view was observedFIGURE 4 | Comparison of (a) temperature and (b) efficiency degradation.FIGURE 5 | Isc–Voc plots of the (a) 35° and (b) two-axis tracking samples.4 of 7 Solar RRL, 2025 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenseat a high magnification, an increase in the emission current den-sity was detected. Therefore, the highly resistive phase wascharged and appeared bright owing to the surface-charge effect(Figure S6). This phase appeared on the light-irradiatedhole-transport-layer side; no such phase detected on the back sidethat contained the Ag electrode. Although the reaction betweenthe metal electrode and halogen is one of the factors of deterio-ration [11], the results obtained in our study indicate the heat-induced phase transition as the primary degradation mechanismof or samples.To corroborate this degradation mechanism, we used a sealed,unexposed sample heated on a hot plate under light irradiationin an indoor solar simulator to reproduce outdoor degradation.After heating for 24 h on a hot plate at 70°C under 1 sun irra-diation in air, spots were observed under an optical microscope(Mean.Value = 25.79; Figure 8a). In addition, cross-sectionalSEM images revealed the presence of a high-resistance phasein the perovskite layer similar to that observed in the outdoor-degraded sample (Figure 8b). Table 1 summarizes the resultsobtained by heating with and without light irradiation underopen-circuit and short-circuit conditions (Figure S7), providinginsights into the effects of various parameters. Evidently, a high-resistance phase was observed in all the heated samples. Theratio of spots under an optical microscope varied as: dark/open< light/open < dark/short << light/short. This result suggeststhat this mode of degradation is affected by—temperature,FIGURE 6 | OM images of the (a) initial and (b) outdoor-degraded samples.FIGURE 7 | Cross-sectional image of the outdoor-degraded sample.FIGURE 8 | (a) OM and (b) cross-sectional SEM images of the indoor-treated samples.TABLE 1 | Summary of indoor testing.Indoor OutdoorHeat 70°C up to 90°CLight w/o w/o 1 sun 1sun up to 1.2 sunCircuit open short open short IV tracingSEM (PhaseChange)✓ ✓ ✓ ✓ ✓OM (Mean.Value)14.57 17.65 16.18 25.79 29.625 of 7 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licenselight, and the flow of electrons. This high-resistance phase maybe indexed as the δ phase that transforms from the α phase.However, no δ-phase diffraction peak was visible in the X-ray diffraction (XRD) profiles of the degraded samples(Figure S8). This absence of δ-phase diffraction peaks may beattributed to the low volume ratio or low crystallinity of thesehigh-resistance phases. Analytical methods, such as transmis-sion electron microscope (TEM) diffraction, for detecting crys-tallographic high-resistance phases in minute areas will bediscussed in the future.3 | ConclusionIn this study, we investigated the long-term outdoor performanceand degradation mechanisms of inverted-structure PSCs over a 3-year period. The results revealed significant efficiency losses, par-ticularly during the summer months; these losses were correlatedwith the increased surface temperatures that exceeded 50°C, andthe estimated cell temperature occasionally approached 90°C.This thermal stress led to the formation of a high-resistancephase within the perovskite layer, primarily on the NiOx side(confirmed by OM and cross-sectional SEM observations). Acomparative analysis of the I–V curve tracing and MPPT meas-urements suggested that compared to the former, the MPPTmethod resulted in slightly less degradation, potentially owingto reduced thermal or electrical stress. The reproduction ofthe degradation phenomena under controlled indoor heatingand illumination conditions confirmed that thermal and photo-induced stress, combined with charge carrier flow, acceleratedthe formation of high-resistance regions. To prevent such phasetransitions, the NiOx surface can be modified by a more stabletreatment [24, 25] or by incorporating trivalent cation speciesinto the B sites [26].Overall, our findings emphasize the importance of thermal man-agement for achieving long-term stability of PSCs, particularlyunder real-world outdoor conditions. Future investigationsshould be focused on improving the thermal stability throughmaterial engineering and module design to suppress heat-induced phase transitions and extend the operational lifetimeof PSCs.4 | Experimental Methods4.1 | Device FabricationInverted structured glass/ITO/NiOx/perovskite/C60/BCP/AgPSCs that were used for the outdoor measurements were fabri-cated at the National Institute for Materials Science, Japan. TheNiOx surface was treated with MeO-2PACz, and the MA/Br-freeperovskite layer was passivated with piperazine dihydride.Detailed information on the fabricated PSCs is reported else-where [6, 27]. Devices were prepared on 5 cm2 substrates con-taining four individual cells, each with a light-receiving areaof 1.26 cm2. The cells were encapsulated on both sides using glassand epoxy resins. For outdoor deployment, copper wires weresoldered to the electrodes, followed by secondary sealing usinga resin and adhesive (Figure 9).4.2 | Indoor TestingThe initial and postdegradation device performances were charac-terized by QEmeasurements (Bunkokeiki CEP-25NLT) and undersimulated AM1.5 solar illumination (EKO Instruments). For QEand current density–voltage (J–V ) characterization, a 1-cm2shadow mask was applied, and all the measurements were per-formed at a controlled temperature of 25°C. The indoor durabilitytesting of the sealed cells conformed to ISOS-L L2 or L3 (asreported by Khadka et al. [6]). Accelerated degradation tests wereconducted using a solar simulator and with substrate heating.Heating the cells to approximately 70°C successfully reproducedthe degradation behavior observed under outdoor conditions.The surface and cross-sectional morphologies were analyzed usingOM (KEYENCE VHX-500F) and field-emission SEM (HITACHISU3480). Crystal structure analysis was conducted using thin-filmXRD (Bruker AXS D8 Discover) with a grazing incidence angle of1.2° and a 2θ scan range of 10°–50°. Prior to the XRD measure-ments, the Ag electrode was removed using Kapton tape.4.3 | Outdoor TestingThe outdoor performance was evaluated using two installationmethods: a fixed south-facing tilt at 35° and a two-axis trackingFIGURE 9 | (a) Schematic and (b) photograph of outdoor-tested PSC samples.6 of 7 Solar RRL, 2025 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensesystem (EKO Instruments). The incident solar irradiance wasmeasured using a dedicated pyranometer. Performance datawere obtained using both the I–V curve tracing and MPPT meth-ods. In the I–V method, I–V sweeps were performed every 2 minbetween 4:00 and 20:00, with a sweep duration of 1 s, and thedevice remained in an open-circuit state between the measure-ments. In contrast, the MPPT method continuously tracked themaximum power point using a hill-climbing algorithm, with I–Vsweeps conducted at 11:00, 12:00, and 13:00 JST each day; thesame sweep duration of 1 s was maintained for both directions.In addition, the solar-cell performance at 12:00 JST on clear dayswas compared with the daily average efficiency. In addition tostandard meteorological data, such as ambient temperature, pre-cipitation, wind speed, and atmospheric pressure, the surfacetemperature of the encapsulated solar cell was monitored byplacing a thermocouple in direct contact with the encapsulantsurface. The samples for this outdoor test were installed at a fixedangle of 35° or on a two-axis tracking system, and MPPT controlwas performed without temperature control in compliance withthe ISOS-O O-3 evaluation level.AcknowledgmentsThis work was supported by the Priority Research Project of Tokyo CityUniversity, Japan.Conflicts of InterestThe authors declare no conflicts of interest.Data Availability StatementThe data that support the findings of this study are available from thecorresponding author upon reasonable request.References1. T. Miyasaka, A. Kojima, K. Teshima, and Y. Shirai, Journal of theAmerican Chemical Society 131 (2009): 6050.2. N. G. Park, Materials Today 18 (2015): 65.3. M. A. Green, E. D. Dunlop, M. Yoshita, et al., Progress in Photovoltaics:Research and Applications 33 (2025): 3.4. S. Liu, V. P. Biju, NPG Asia Materials, 15 (2023): 27, https://doi.org/10.1038/s41427-023-00474-z.5. K. Miyano, N. Tripathi, M. Yanagida, and Y. Shirai, Accounts ofChemical Research Journal 49 (2016): 303.6. D. B. Khadka, Y. Shirai, M. Yanagida, et al., Nature Communication 15(2024): 882, https://doi.org/10.1038/s41467-024-45228-9.7. T. Matsui, T. Yamamoto, T. Nishihara, et al., Advanced Materials 31(2019): 1806823, https://doi.org/10.1002/adma.201806823.8. J. Zhuang, J. Wang, and F. Yan, Nano-Micro Letters 15, 2023: 84,https://doi.org/10.1007/s40820-023-01046-0.9. D. B. Khadka, Y. Shirai, M. Yanagida, and K. Miyano, ACS AppliedEnergy Materials 4 (2021): 11121.10. D. B. Khadka, M. Yanagida, and Y. Shirai, Solar Energy Materials andSolar Cells 281 (2025): 113319, https://doi.org/10.1016/j.solmat.2024.113319.11. S. Baumann, G. E. Eperon, A. Virtuani, et al., Energy EnvironmentalScience 17 (2024): 7566.12. M. Jošt, B. Lipovšek, B. Glažar, et al., Advanced Energy Materials 10(2020): 2000454, https://doi.org/10.1002/aenm.202000454.13. M. Konagai, H. Okawa, R. Ishikawa, M. Yanagida, and Y. Shirai, Proc.of the 34th International Photovoltaic Science and Engineering Conf.(PVSEC-34), (2023), 222.14. V. Paraskeva, M. Norton, A. Livera, et al., ACS Energy Letters 9 (2024):5081.15. M. Khenkin, H. Köbler, M. Remec, et al., Energy EnvironmentalScience 17 (2023): 602.16. G. Hodes, Science 342 (2013), 317, https://doi.org/10.1126/science.1245473.17. M. Saliba, T. Matsui, K. Domanski, et al., Science 2016 (1979): 354.18. T. Haeger, R. Heiderhoff, and T. Riedl, Journal of Materials ChemistryC 8 (2020): 14289.19. M. Nakamura, I. Takenaka, T. Mabuchi, et al., ACS Applied EnergyMaterials 5 (2022): 10409.20. M. B. Islam, M. Yanagida, Y. Shirai, Y. Nabetani, and K. Miyano, ACSOmega 2 (2017): 2291.21. L. Cojocaru, S. Uchida, K. Tamaki, et al., Science Reports 7 (2017): 1.22. T. Moot, J. B. Patel, G. McAndrews, et al., ACS Energy Letters 6 (2021):2038.23. S. Charan, M. Konagai, and K. Takahashi, Journal of Applied Physics50 (1979): 963.24. J. Mohanraj, B. Samanta, O. Almora, et al., ACS Applied MaterialInterfaces 16 (2024): 42835.25. S. Zhang, F. Ye, X. Wang, et al., Science 380 (2023): 404.26. R. Nie, Y. Dai, R. Wang, et al., Nature Communication 16 (2025): 1.27. M. Yanagida, T. Nakamura, T. Yoshida, D. B. Khadka, Y. Shirai, andK. Miyano, Japan Journal of Applied Physics 62 (2023): SK1054.Supporting InformationAdditional supporting information can be found online in the SupportingInformation section. Supporting Fig. 1: Initial PSC performance mea-sured via indoor tests. Supporting Fig. 2: Transmittance of the NiOx layer.Supporting Fig. 3: Variations in Rsh and Rs of the degradedPSCs. Supporting Fig. 4: Comparison of humidity and efficiency degra-dation. Supporting Fig. 5: Voc as a function of surface temperature.Supporting Fig. 5: Comparison of the same field of cross-sectionalSEM images at different magnifications. Supporting Fig. 7: Cross-sectional SEM images (a–c) and OM images (d–f ) of dark/open, light/open,and dark/short conditions. Supporting Fig. 8: XRD patterns of theoutdoor-degraded sample. Supporting Table 1: Initial PSC performanceparameters.7 of 7 2367198x, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/solr.202500524 by National Institute For, Wiley Online Library on [10/10/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttps://doi.org/10.1038/s41427-023-00474-zhttps://doi.org/10.1038/s41427-023-00474-zhttps://doi.org/10.1038/s41467-024-45228-9https://doi.org/10.1002/adma.201806823https://doi.org/10.1007/s40820-023-01046-0https://doi.org/10.1016/j.solmat.2024.113319https://doi.org/10.1016/j.solmat.2024.113319https://doi.org/10.1002/aenm.202000454https://doi.org/10.1126/science.1245473https://doi.org/10.1126/science.1245473 Outdoor Performance and Thermally Accelerated Degradation of Inverted Perovskite Solar Cells 1. Introduction 2. Results and Discussion 2.1. Long-Term Outdoor Degradation Behavior 2.2. Comparison between I-V Curve Tracing and MPPT 2.3. Temperature Dependance 2.4. Estimation of Cell Temperature 2.5. Microscopic and Structural Changes 3. Conclusion 4. Experimental Methods 4.1. Device Fabrication 4.2. Indoor Testing 4.3. Outdoor Testing