Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Nature Communications volume 16, Article number: 10355 (2025)
4924
5
Metrics details
Reverse bias stability remains a critical challenge for inverted perovskite solar cells (PSCs). While self-assembled monolayers (SAMs) boost efficiency, their low breakdown voltages limit device reliability. Thick PTAA layer improves breakdown voltage but suffers from poor wettability and efficiency loss, with unclear effects on device reverse bias stability. Here, we use electroluminescence mapping to reveal the critical role of hole transport layer (HTL) uniformity in affecting device reverse bias stability, and poor uniformity of current HTLs causes spatial heterogeneity that is not able to block electron injection and leads to device breakdown under reverse bias. Based on our study, we develop a polymeric Poly-PhPACz HTL with high conductivity and good wettability, achieving a breakdown voltage comparable to PTAA while maintaining high efficiencies across varying thicknesses. Ambient blade-coated Poly-PhPACz PSCs achieve 26.1% efficiency and retain 92% performance after 1,800 hours of light soaking. Further optimization yields a high breakdown voltage of −14.3 V without sacrificing efficiency, offering a promising pathway for stable PSCs.
Inverted perovskite solar cells (PSCs) with a p-i-n architecture have emerged as a highly promising candidate for photovoltaic applications owing to their low manufacturing cost, solution processability, mechanical flexibility, and lightweight characteristics1,2,3,4,5. Notably, the recent advancement of self-assembled monolayer (SAM)-based hole-transporting layers (HTLs) has propelled their power conversion efficiencies (PCEs) to levels competitive with conventional silicon photovoltaics6,7,8,9,10,11. While pursuing high PCEs continues to be a primary research focus, device stability has concurrently emerged as an equally crucial consideration12,13. Significant improvements in the operational stability of laboratory-scale PSCs have been achieved through various strategies, including ink formulation optimization, interface passivation, device structure engineering, and advanced encapsulation techniques14,15,16,17,18,19.
Despite these advancements, the reverse bias stability of inverted PSCs has received comparatively limited attention, despite its critical impact on device longevity20,21,22,23. Reverse bias instability arises when mismatched current generation occurs in serially connected photovoltaic module—such as when shaded or underperforming cells are forced to accommodate the higher current of neighboring cells24. Under reverse bias, the affected cell dissipates power as heat, leading to localized temperature rise and irreversible degradation25. Unlike commercial silicon photovoltaics, which typically exhibit a well-defined reverse breakdown voltage (Vrb) around −15 V, PSCs demonstrate significantly lower Vrb values (ranging from −0.5 V to −4.0 V)26,27. It can be anticipated that the Vrb of PSCs is closely linked to interface/electrode stability, where reverse bias is applied, and weak interfaces or active electrodes accelerate device breakdown failure. To bridge this performance gap, developing robust and uniform interfacial layers is imperative for enhancing the Vrb of PSCs without compromising their photovoltaic efficiency. Addressing this challenge is essential for ensuring the commercial viability of perovskite solar modules.
Several approaches have been explored to improve the reverse bias stability of inverted PSCs, including optimizing charge transport layers and employing more robust electrode materials. For instance, Huang et al. improved reverse bias stability in both small-area cells and modules by redesigning the electron transporters. By replacing the conventional fullerene (C60)/bathocuproine (BCP) bilayer with a four-layer stack of LiF/C60/atomic-layer-deposited SnO2/indium tin oxide (ITO), they achieved a Vrb of approximately −15 V—comparable to silicon photovoltaics—while maintaining high efficiency27. However, this approach relies on time-consuming atomic layer deposition and magnetron sputtering, significantly complicating device fabrication.
Recently, Ginger et al. employed a thick PTAA layer (over 30 nm) as the HTL, achieving a Vrb of −7.6 V with an average PCE of 14.17%. In contrast, devices using Me-2PACz SAM as the HTL exhibited a substantially lower Vrb (−1.03 V). Further replacing the silver electrode with gold increased the Vrb of thick-PTAA-based PSCs to ~−15 V, confirming the efficacy of interface/electrode optimization in enhancing device reverse bias tolerance26. However, PTAA’s intrinsic hydrophobicity poses challenges for large-scale perovskite deposition if no suitable surface treatment is applied, particularly at an increased thickness28,29,30,31,32, and also gold electrodes are prohibitively expensive for commercial applications.
Although increasing the HTL thickness (SAMs or PTAA) via high-concentration HTL solutions might seem viable, prior work reveals a critical trade-off: devices employing conventional small-molecule SAMs suffer from rapid PCE degradation as SAM thickness increases due to the insulating nature of SAMs33,34,35,36, and thicker PTAA causes poor wettability for large-area perovskite deposition31,32. Thus, developing scalable and cost-effective strategies that simultaneously improve Vrb and maintain high PCEs remains a key challenge for improving the reverse bias stability of p-i-n PSCs. Meanwhile, the role of HTLs in affecting device reverse bias stability has not been thoroughly investigated, and there is a lack of direct experimental evidence to elucidate the impacts of HTL properties on reverse bias stability.
To shed light on these issues, by using electroluminescence (EL) mapping, we directly visualized the key differences in film uniformity and charge injection behavior of current HTLs that significantly affect device reverse bias stability. Our study unveils the limitation of current SAM HTLs (Fig. 1a) in interface uniformity that generates spatial heterogeneity, which is not able to block electron injection and ultimately leads to device breakdown under reverse bias. Thick PTAA HTL helps to mitigate this detrimental issue, but at a compromise of both device PCEs and scalable processability. To this end, we designed Poly-PhPACz, a polymeric hole transport material with both improved processability and stability. Unlike conventional SAM-based HTLs such as Me-4PACz and PhPACz, Poly-PhPACz exhibits high uniformity, thickness insensitivity, and interfacial robustness-critical for scalable fabrication. Remarkably, devices incorporating Poly-PhPACz achieve a Vrb of beyond −8.0 V, with optimized configurations reaching −14.3 V, rivaling silicon photovoltaics. This enhancement stems from Poly-PhPACz’s high film uniformity that significantly suppresses the formation of localized heterogeneity and thus charge injection paths. Meanwhile, ambient blade-coated PSCs employing Poly-PhPACz deliver a high PCE of 26.1% (25.85% certified), significantly outperforming devices with PhPACz (21.6%). Furthermore, these devices demonstrate good operational stability, retaining >92% of their initial PCE after 1800 h under continuous one-sun illumination (100 mW cm−2). By addressing both reverse bias instability and scalability challenges, this work represents a critical step toward the commercialization of perovskite photovoltaics, offering a viable pathway to bridge the performance-reliability gap in next-generation solar cells.
a Chemical structures of the HTLs studied in this work. b Dark J–V curves of the PSCs to determine the cell Vrb. c–g Normalized EL mapping (c–f) and EL intensity distributions (g) of the PSCs based on different HTLs. a.u. arbitrary units.
We systematically investigated the breakdown behavior of p-i-n structured PSCs with a typical architecture of glass/ITO/HTL/MA0.7FA0.3PbI3/C60/BCP/Cu (MA: methylammonium; FA: formamidinium), where HTLs include current benchmark SAMs (e.g., Me-4PACz and PhPACz) and PTAA. Both HTLs and perovskite layers were blade-coated under ambient conditions (25 °C, 30–40% RH). These PSCs exhibited comparable PCEs as previously reported (Supplementary Fig. 1 and Supplementary Table 1)33,34. Reverse bias testing was performed by applying a positive voltage to the Cu back electrode while biasing the ITO side negatively, enabling electron injection from ITO and hole injection from Cu. More testing details can be found in the characterization section and Supplementary Note. Increasing reverse bias voltage induced a dramatic current surge and finally led to irreversible device breakdown (Supplementary Movie 1), revealing critical breakdown characteristics (Fig. 1b and Supplementary Fig. 2). Devices without any HTL exhibited a Vrb of −1.2 V, while incorporation of Me-4PACz, PhPACz, or other SAMs enhanced Vrb to a range of −3.5 to −4.7 V, respectively, demonstrating the crucial role of HTLs in mitigating reverse bias degradation and consistent with the previous report23,26,37.
In addition to sequential deposition of SAM HTL and perovskite layer, we further examined the reverse bias stability based on the co-deposition of SAM and perovskite layers38. This co-deposition approach involves incorporating the SAM directly into the perovskite precursor solution, allowing SAM molecules to adsorb onto the ITO substrate during the solution drying and perovskite crystallization process. Such methodology has been proposed to streamline the fabrication of perovskite photovoltaic devices. Given the crucial influence of HTLs on device stability under reverse bias conditions, we specifically focused on examining their reverse bias performance characteristics and evaluated their suitability for scalable fabrication. Our study shows that the PSC prepared by the co-deposition method of PhPACz yielded a substantially lower Vrb of −2.9 V compared to sequential deposition. While this co-deposition strategy offers processing advantages, its inadequate reverse bias stability (as evidenced by the low Vrb) may render it unsuitable for commercial applications. In contrast, devices employing PTAA (24 nm) as the HTL demonstrated significantly enhanced reverse bias tolerance with a Vrb of −8.3 V, confirming the apparent advantage of conductive polymers over molecular HTLs.
Understanding how HTLs influence device reverse bias behavior remains a significant challenge in perovskite photovoltaics, as direct experimental approaches or conclusive evidence to elucidate the underlying mechanism are still lacking. This issue is particularly pronounced in SAM-based PSCs, where the ultrathin nature of SAM HTLs makes them highly difficult to characterize. Currently, even fundamental aspects such as the stacking configuration and morphology of SAM HTLs on conductive substrates remain debated and unresolved39. To address this knowledge gap, we propose employing EL mapping as a powerful tool to investigate the role of different HTLs under bias conditions. This approach will provide critical insights into how HTLs govern device stability when charges are injected, ultimately advancing our understanding of their functional mechanisms. Unlike conventional photoluminescence (PL) mapping, which is limited to probing the photoactive layer alone, EL mapping enables the characterization of solar cells under conditions that closely resemble actual device operation. Moreover, PL mapping is often compromised by artifacts arising from lateral carrier diffusion near the excitation spot. In contrast, EL mapping serves as a complementary technique that not only quantifies the luminescence flux of full devices but also avoids the confounding effects inherent to confocal PL measurements11,40. Supplementary Fig. 3 illustrates our EL mapping setup, where EL emission was induced by applying a forward bias.
In this investigation, we conducted high-resolution EL mapping at a forward bias that injected comparable current density to device JSC obtained under one-sun illuminance, employing a 120 × 120 µm scanning area with 160 nm pixel resolution41. Figure 1c–g presents the normalized 2D EL intensity maps and corresponding 1D intensity distributions. The EL intensity profiles are also extracted to better show the fluctuation of emission intensity (Supplementary Fig. 4). The PSCs incorporating a separate PhPACz layer demonstrated markedly non-uniform EL emission (Fig. 1c), as evidenced by numerous discrete luminous hotspots, revealing significant limitations in hole transport homogeneity. Such spatial heterogeneity in charge injection becomes particularly pronounced under high bias conditions, ultimately providing preferential pathways for reverse bias breakdown. Although EL emission and reverse breakdown occur under distinct bias conditions, the observed spatial heterogeneity in EL mapping remains as localized change injection paths affecting device stability across different operational biases.
Notably, devices fabricated through the co-deposition approach exhibited further degradation in both EL intensity and spatial uniformity compared to sequentially deposited HTL/perovskite structures. This observation strongly suggests incomplete and non-uniform adsorption of SAM molecules onto ITO during the solution drying process, highlighting a critical challenge in the co-deposition methodology that should be given serious consideration when evaluating its potential in scalable manufacturing. These results suggest the intrinsic disadvantage of SAM-type HTLs in achieving scalable and uniform deposition.
Regarding the thick-PTAA (24 nm)-based device, it demonstrated enhanced EL uniformity (Fig. 1e) and a narrower intensity distribution (Fig. 1g) compared to SAM counterparts. This observation highlights the improved performance of thick polymeric HTLs over SAMs in facilitating uniform charge transport and injection. The improved homogeneity reduces localized failure spots, thereby suppressing the formation of preferential charge injection pathways that contribute to device degradation under reverse bias conditions. However, residual dark spots remain observable within the field of view, likely attributable to interfacial traps acting as non-radiative recombination centers.
The EL mapping analysis conclusively establishes that HTL uniformity is a critical performance metric influencing device stability under reverse bias. Given the inherent limitations of SAM-based HTLs—particularly their poor uniformity and ultrathin deposition on ITO due to their insulating nature—the development of conjugated polymeric alternatives such as PTAA becomes essential. Such materials enable efficient hole transport even at greater thicknesses. Nevertheless, PTAA presents certain challenges: its hydrophobic nature complicates large-area perovskite film deposition, while additional surface treatments raise production costs. Furthermore, the absence of functional groups capable of robust adsorption onto ITO substrates compromises interfacial stability during device operation.
The design of HTLs should comprehensively address the limitations of current HTL materials while retaining their advantageous properties. Conventional SAMs are intrinsically insulating small molecules, and they have to form an ultrathin monolayer to reduce charge-transport resistance. Once their layer thickness increases, the resistance increases significantly and deteriorates hole extraction efficiency33. To overcome these challenges, we developed a conjugated polymer, Poly-PhPACz, synthesized through polymerization of a SAM molecule (Fig. 1a). The conjugation formed along the polymer backbone can significantly facilitate the intra- and inter-molecular charge transports, forming good coverage on ITO and reducing current leakage. Meanwhile, Poly-PhPACz also possesses phosphorous acid side-chains, enabling it to bind to ITO like SAMs and stabilize the interface. Detailed synthesis is provided in Supplementary Information. This polymeric material maintains high electrical conductivity while exhibiting strong chemical binding to transparent conductive substrates. As demonstrated in Fig. 1f, PSCs employing Poly-PhPACz (20 nm) as the HTL display high EL uniformity, and achieve the highest EL intensity and narrowest EL distribution among the four cases (Fig. 1g). The complete absence of observable hotspots, coupled with uniform EL emission across the entire field of view, indicates that the homogeneous deposition of Poly-PhPACz effectively transports injected holes to perovskite layer and excites the EL emission. Such a spatial homogeneity is able to block electron injection under reverse bias. Consequently, Poly-PhPACz-based devices achieve a remarkable Vrb of −8.0 V (Fig. 1b and Supplementary Movie 2), comparable to devices incorporating thick PTAA layers. This indicates that the improvement of reverse bias stability based on polymeric HTLs relies on both their better uniformity and higher conductivity compared to SAM counterparts. Conjugated polymers like PTAA and Poly-PhPACz can form relatively thicker HTLs with significantly improved uniformity and electron-blocking capacity. This shows the intrinsic advantages of conjugated HTL polymers over SAMs in terms of reverse bias stability. We also studied the temperature change of the Poly-PhPACz-based PSC under reverse bias, and it was found that the device exhibited an irreversible thermal breakdown behavior (Supplementary Fig. 5)42. The leakage current analysis of the PSCs based on three HTLs was further carried out by continuously biasing the PSCs at −2 V under dark conditions. The current changes were monitored (Supplementary Fig. 6), and the Poly-PhPACz-based device showed the highest resistance to reverse bias. These findings collectively establish a fundamental correlation between interfacial uniformity and device breakdown behavior under reverse bias conditions, providing insights into the formation mechanisms of spatial heterogeneity and their impact on charge injection dynamics.
To investigate the dependence of PCE and Vrb on HTL thickness, we fabricated devices with varying Poly-PhPACz thicknesses. As illustrated in Fig. 2a, increasing the Poly-PhPACz thickness leads to higher Vrb values, with the maximum Vrb reaching −8.4 V at a thickness of 63 nm, at which the corresponding PSC can still maintain a decent PCE of over 20%. Furthermore, PSCs with thicker Poly-PhPACz layers demonstrate a narrower distribution of Vrb values, confirming that increased HTL thickness improves device reproducibility under reverse bias. Critically, the device still achieved a high PCE of 24.2% at a Poly-PhPACz layer thickness of 20 nm (its optimal PCE will be discussed in the following section), while maintaining a competitive Vrb of approximately −8.0 V compared to that of a 24 nm-thick PTAA-based device. However, the PCE of the PTAA-based device dropped significantly to 20.5% and 18.1% at the thicknesses of 24 nm and 31 nm, respectively (Supplementary Fig. 7), highlighting the improved performance of Poly-PhPACz in terms of a wide processing window. Furthermore, PTAA exhibited inferior film wettability compared to Poly-PhPACz, as evidenced by contact angle measurements (Fig. 2b). The water contact angle for Poly-PhPACz was 52.7°, significantly lower than PTAA’s 79.5°. The perovskite ink also shows a much smaller contact angle of 13.8° on Poly-PhPACz than that of 40.1° on PTAA (Supplementary Fig. 8), indicating better perovskite ink spreading on Poly-PhPACz. This enhanced wettability improves HTL uniformity, facilitating the formation of a more homogeneous perovskite layer and compact interfacial contact with the underlying HTL.
a Dependence of device PCE/Vrb on the thickness of Poly-PhPACz HTL. b Water contact angles of the ITO substrates coated with PTAA or Poly-PhPACz. c In 3d XPS spectra of bare ITO and the ITO substrates coated with PhPACz or Poly-PhPACz. d TRPL and steady-state PL (inset) spectra of the perovskite films coated on PhPACz and Poly-PhPACz glass substrates. e Backward J–V curves of PhPACz and Poly-PhPACz devices (the inset shows stabilized powder output curves). f EQE curves of the optimized devices from 300 nm to 900 nm. a.u. arbitrary units.
To elucidate the relationship between HTL properties and device performance, we further systematically compared the optoelectronic characteristics of Poly-PhPACz and its monomeric counterpart (PhPACz). Ultraviolet-visible (UV-vis) spectra revealed a distinct red-shift absorption of Poly-PhPACz thin film (Supplementary Fig. 9), consistent with the formation of an extended π-conjugated polymer backbone. Ultraviolet photoelectron spectroscopy (UPS) measurements demonstrated significant differences in electronic structures (Supplementary Fig. 10): Poly-PhPACz exhibited a shallower valence band maximum of −5.05 eV than that of PhPACz (−5.74 eV), which was attributed to electron delocalization along the conjugated polymer backbone. X-ray photoelectron spectroscopy (XPS) analysis of In 3d provided further insight into interfacial interactions. Both HTLs induced substantial peak shifts relative to bare ITO, with Poly-PhPACz showing a more pronounced shift by 1.3 eV (1.0 eV for PhPACz, Fig. 2c), indicating its stronger chemical bonding at the ITO interface. This enhanced interfacial interaction likely improves the operational stability of Poly-PhPACz-based devices.
PL characterization of the blade-coated perovskite layers reveals significant differences in charge recombination dynamics at the HTL/perovskite interfaces (Fig. 2d), with steady-state PL measurements demonstrating a 1.36-fold enhancement in emission intensity for Poly-PhPACz compared to PhPACz counterpart. Time-resolved PL analysis further verified this improvement, showing extended average carrier lifetimes of 2655.0 ns for Poly-PhPACz versus 2051.9 ns for PhPACz (Supplementary Table 2), collectively indicating the suppression of non-radiative recombination pathways at the Poly-PhPACz/perovskite interface, which is attributed to the passivation effect of Poly-PhPACz to perovskite interfacial defects. In addition, the scanning electron microscopy images and X-ray diffraction patterns of the perovskite films deposited on PhPACz and Poly-PhPACz were measured, and the results showed no significant difference in grain size distribution or diffraction intensity (Supplementary Fig. 11). The passivation effect is attributed to a much higher amount of phosphorous acid groups in the thicker Poly-PhPACz layer. Although PhPACz also has a phosphorous acid group, most of them have to bind to ITO and self-assemble into an ultrathin SAM layer. On the contrary, the thick Poly-PhPACz layer has more phosphorous acid groups. On the one hand, this high density of phosphorous acid groups binds to ITO and stabilizes the interface. On the other hand, the remaining ones tune the HTL surface wettability and interact with the perovskite layer, promoting ink spreading and passivating interface defects, respectively.
Efficient hole transport requires high conductance of HTLs to avoid charge recombination. To evaluate the conductivity of the two HTMs, conductive atomic force microscopy (c-AFM) measurements were conducted. Thick Poly-PhPACz film demonstrated a higher average current (3.68 nA) compared to PhPACz (2.75 nA) (Supplementary Fig. 12), explaining the independence of device PCEs on Poly-PhPACz thickness. Meanwhile, I-V curves of the glass/ITO/HTL/Ag devices were measured to evaluate HTL conductivity (Supplementary Fig. 13), and Poly-PhPACz shows the highest conductivity compared with PhPACz and PTAA. Hole mobilities of HTLs were also investigated via the space-charge-limited-current (SCLC) method, and Poly-PhPACz exhibits the highest hole mobility of 4.6 × 10−5 cm−2 V−1 s−1 (Supplementary Fig. 14). Furthermore, to assess the impact of different HTLs on carrier transport for complete PSC devices, the photo-induced charge extraction linearly increasing voltage (photo-CELIV) method was utilized to determine the average charge mobility (µavg). The results of µavg values are 1.46 × 10−3 and 9.13 × 10−4 cm2 V−1 s−1 for Poly-PhPACz and PhPACz-based PSCs (Supplementary Fig. 15). Since the two types of PSC devices have the same structure except for the HTL, the photo-CELIV results suggest that Poly-PhPACz plays a key role in improving charge extraction.
In terms of PSC photovoltaic performance, J–V characterization measured under AM 1.5G one-sun illuminance (100 mW cm−2) revealed significant performance enhancement when using Poly-PhPACz compared to its molecular counterpart (Fig. 2e and Supplementary Fig. 16). PhPACz-based devices showed a decent PCE of 21.6%, along with an open-circuit voltage (VOC) of 1.14 V, a short-circuit current density (JSC) of 24.5 mA cm−2, and a fill factor (FF) of 77.4%. Poly-PhPACz devices achieved a remarkable PCE of 26.1% with improved parameters across all metrics (VOC = 1.18 V, JSC = 26.2 mA cm−2, FF = 84.3%). A stabilized PCE of 25.4% was obtained by tracking the current output biased at a fix voltage for 600 s, which is much higher than that of PhPACz (20.4%, inset of Fig. 2e). One of the best-performing Poly-PhPACz devices was validated by an independent solar cell-accredited laboratory (the National Photovoltaic Product Quality Inspection and Testing Center, China) for certification, where a PCE of 25.85% was confirmed (Supplementary Fig. 17). The statistical device parameters are detailed in Supplementary Fig. 18 and Supplementary Table 3, and the Poly-PhPACz-based PSCs also showed better reproducibility. External quantum efficiency (EQE) measurements (Fig. 2f) consistently demonstrated higher photon-to-current conversion efficiency across the entire spectral range for Poly-PhPACz devices. We also fabricated 1-cm2 devices, and their EL mapping and J–V curves are shown in Supplementary Fig. 19. Compared to PTAA, the EL emission of the Poly-PhPACz device was more uniform, while the PTAA device showed a few dispersed dark spots.
The charge carrier lifetimes derived from transient photovoltage (TPV) measurements (Fig. 3a) reveal that the device utilizing Poly-PhPACz exhibits a longer TPV lifetime of 4.46 μs compared to the PhPACz device (3.30 μs). Additionally, Supplementary Fig. 20a presents the transient photocurrent (TPC) results and the corresponding charge extraction times. The device based on Poly-PhPACz demonstrates a shorter charge extraction time of 0.19 μs, whereas the device based on PhPACz shows a longer duration of 0.21 μs. These findings further confirm that employing Poly-PhPACz as the HTL facilitates rapid charge carrier extraction. These device photophysics characterizations underscore that the Poly-PhPACz device offers several advantages over PhPACz, including reduced charge recombination and expedited charge extraction. The light intensity-dependent VOC measurements allowed for the determination of the ideality factors, which were fitted to be 1.14 and 1.27 for the Poly-PhPACz- and PhPACz-based PSCs (Fig. 3b). The lower ideality factor observed in these devices is ascribed to a reduction in charge traps (Supplementary Fig. 20b)43. To further confirm the passivation effect of Poly-PhPACz in PSCs, we conducted trap density of states (tDOS) measurements on the two groups of PSCs (Fig. 3c)44. The tDOS results revealed that devices based on Poly-PhPACz exhibited a lower trap density, particularly in the shallower depth region from 0.10 eV to 0.35 eV. To further investigate the spatial reduction of trap density, we employed the drive-level capacitance profiling (DLCP) method (Fig. 3d, 10 kHz)45. The results showed that Poly-PhPACz effectively reduced the trap density near the HTL interface.
TPV curves (a), light intensity dependence of VOC (b), tDOS spectra (c), DLCP curves (d), and MPPT test results (e) under simulated AM 1.5G one-sun illuminance (100 mW cm−2) in air of the PSCs based on different HTLs. a.u. arbitrary units.
In order to further investigate the photostability of the encapsulated devices, we conducted a maximum power point tracking (MPPT) test under simulated AM 1.5G one-sun illumination (100 mW cm−2). The cell temperature was measured to be approximately 40 °C, with a relative humidity of about ~30–40%. As shown in Fig. 3e, devices based on Poly-PhPACz exhibited significantly better operational stability, retaining over 92% of their initial PCE after more than 1800 h of light soaking, while the PCE of the PhPACz-based device degraded to 11.7% after 800 h. This clearly demonstrates the effectiveness of Poly-PhPACz in enhancing the photostability of PSCs.
As shown in Supplementary Movies 1 and 2, increasing the reverse bias also damaged the back Cu electrode that was biased positively. To further enhance the Vrb value based on Poly-PhPACz, we fixed at 20-nm Poly-PhPACz HTL and continued to optimize device back electrodes. We tried the Ag electrode first, and the resultant devices suffered from lower Vrb values, as shown in Supplementary Fig. 21. The Ag electrode has been studied to diffuse to the perovskite layer and form AgI, which causes the device shunting and thus reduces Vrb46. In addition, the Ag electrode also tends to be oxidized when it is biased positively under the reverse bias condition. Instead of resorting to complex fabrication processes or costly gold electrodes, we employed a previously reported approach involving a thermally evaporated Cr/Cu double layer to further improve device Vrb47,48. The insertion of Cr can shield the diffusion of the top metal contact to perovskite, and Cr2O3 has also been reported to form with the trace amount of O2 during the thermal evaporation of the Cr layer, thus suppressing the formation of metal iodide species47,48,49,50.
By optimizing the thicknesses of the Cr/Cu double layer (100 nm in total), we effectively increased the Vrb while maintaining a high efficiency with 30 nm Cr and 70 nm Cu (Supplementary Fig. 22). As illustrated in Fig. 4a, the Cr/Cu electrode-based device achieved a higher Vrb of −14.3 V compared to the Cu electrode device, while retaining a PCE of 22.7% (Supplementary Fig. 23 and Supplementary Table 4). Under high reverse bias voltage, the Cr/Cu devices did not experience immediate breakdown like the Cu devices. Instead, micropores slowly increased, which did not render the device completely inoperative (Fig. 4b and Supplementary Movie 3). This gradual change is a key reason for the improved Vrb when using dual-layer electrodes. To assess the long-term reverse bias stability of both device types, a fixed voltage of −2 V was applied, and PCE evolutions were periodically measured. As shown in Fig. 4c and Supplementary Table 5, after 100 h, the PCE of the Cu device decreased from 23.7% to 18.9%, and the FF dropped from 79.0% to 64.8%. In contrast, the efficiency of the Cr/Cu devices decreased from 22.8% to 21.8%, with the FF dropping only slightly from 79.9% to 77.1%. Even after 180 h of continuous reverse bias, the device PCE only declined to 19.8%. Figure 4c, d indicates that the degradation in device performance is mainly due to the decline in FF, with VOC and JSC showing minimal changes. It suggests that the insertion of the Cr layer helps to mitigate the diffusion/drift between Cu and iodide. Compared to previous approaches of using expensive Au electrodes or complicated fabrication procedures, such a simple strategy of employing a double-layer metal electrode has more promising applications in the scalable manufacturing of perovskite photovoltaics.
a Dark J–V curves of PSCs based on Cu and Cr/Cu electrodes (five individual devices for each group). b Device optical images (2 mm by 2 mm) before and after applying reverse bias. J–V curves (AM 1.5G, 100 mW cm−2) of Cu (c) and Cr/Cu (d) devices after being biased at −2 V for different durations.
In summary, this study elucidates the critical role of HTL uniformity in affecting device stability under reverse bias conditions through EL mapping analysis. We demonstrate that the inferior uniformity of SAM-type HTLs generates spatial heterogeneity, ultimately leading to irreversible device breakdown under reverse bias. Increasing PTAA thickness can alleviate this issue at the expense of compromised device efficiency and scalability. To address these challenges, we developed a conductive polymer exhibiting high uniformity, which effectively eliminates localized charge injection pathways and enhances reverse bias stability to levels comparable to thick PTAA layers while maintaining both high device efficiency and process scalability. The resulting ambient blade-coated devices incorporating Poly-PhPACz achieved a champion PCE exceeding 26% and high operational stability. Notably, through simple electrode optimization, we attained a Vrb of approximately −14.3 V. This work not only provides a viable strategy for enhancing the reverse bias stability of PSCs but also represents a significant advancement toward the scalable manufacturing of high-performance perovskite photovoltaics.
BCP, lead iodide (PbI2, 99.999% trace metals), dimethyl sulfoxide (DMSO), L-α-phosphatidylcholine (LP), benzylhydrazine hydrochloride (BHC), and 2-methoxyethanol (2-ME) were purchased from Merck and used without further purification. C60 was purchased from Shenzhen eFlex Inc. MAI, FAI, 4-fluoro-phenyethylammonium iodide (p-F-PEAI), and n-dodecylammonium iodide were purchased from Greatcell Solar Materials. PTAA was purchased from Merck. Me-4PACz, PhPACz, 2PACz, 4PACz, and MeO-2PACz were purchased from Suzhou Liwei New Material Co. Ltd, China. The synthesis details of Poly-PhPACz were provided in the Supplementary Information.
All PSCs were prepared by blade-coating at room temperature inside a fume hood with a relative humidity of 30%–40%. The ITO glass substrates (15 × 15 mm) were cleaned with detergent and then sonicated with deionized water, acetone, and isopropanol, subsequently, and dried overnight in an oven. The ITO substrates were treated with UV-Ozone for 30 min. The HTL solutions (SAMs in methanol, PTAA in toluene, and Poly-PhPACz in methanol:chloroform = 1:1, v/v) were blade-coated onto ITO glass substrates at a speed of 20 mm s−1, followed by thermal annealing at 100 °C for 10 min in air. Their thickness was measured by a surface profiler. A 1.35 M MA0.7FA0.3PbI3 precursor solution was prepared by dissolving the corresponding organic halide salts and PbI2 into 2-ME. 0.83 mg ml−1 n-dodecylammonium iodide, 0.27 mg ml−1 LP, 0.14% v/v MAH2PO2, 1.4 mg ml−1 p-F-PEAI, 0.15 mg ml−1 BHC, and 2.8% v/v DMSO were added as additives. For the co-deposition method, 0.5 mg ml−1 PhPACz was directly added into the perovskite precursor solution. Subsequently, the perovskite precursor solution was blade-coated onto the HTL-covered ITO glass substrates with a gap of 250 μm at a movement speed of 20 mm s−1. The air knife worked at 20 psi during blade-coating. After that, the perovskite films were annealed at 120 °C for 10 min in air to remove the residual solvents. Then, C60 (40 nm, 0.3 Å s−1) and BCP (6 nm, 0.1 Å s−1) were deposited by thermal evaporation. For the back metal electrode, 100 nm Cu (0.1 Å s−1) or a double layer (100 nm in total) of Cr (0.3 Å s−1) and Cu (0.1 Å s−1) were deposited by thermal evaporation. The cells were encapsulated with a cover glass sealed by an epoxy encapsulant.
Reverse bias J–V was acquired using a Keithley 2400 source meter from 2 V to −20 V at a scan rate of 0.1 V s−1 in the dark. Vrb was defined as the reverse bias when the device exhibited irreversible thermal breakdown and the injected current density surged to 100 mA cm−2. The imaging system of EL mapping was conducted using an inverted microscope (Eclipse Ti-U, Nikon) with an Autolab potentiostat (PGSTAT302N). A forward voltage was applied to stimulate luminescence. The EL light from the cell was captured by a CCD camera (Stingray, Allied Vision Technologies) through a 60× objective. A data acquisition card (USB-6281, National Instruments) was employed to synchronize the voltage output from the potentiostat and transistor-transistor logic signals from the camera. The light J–V characteristics of solar devices were measured using a 3A Class solar simulator (BG-LED3A-100S, Class AAA Solar Simulator), and the power of the simulated light was calibrated to 100 mW cm−2 by a silicon reference cell (Newport 91150 V). All devices were measured using a Keithley 2400 source meter with a scan rate of 0.1 V s−1 in the air at room temperature. A metal mask with an aperture (7.4 mm2) aligned with the solar cell area was used for measurements. The stability of encapsulated cells was monitored with a 91 PVKSOLAR MSCLT-1 automatic MPP tracker, and two cells kept working at MPP conditions under simulated AM 1.5G one-sun illumination (100 mW cm−2) in air (relative humidity: 30%–40%). No fan or cooler was used to control the cell temperature, and the temperature of the cells was measured to be ≈40 °C. The EQE spectra were obtained with a Zolix SCS600 solar cell quantum efficiency measurement system equipped with a standard Si cell. c-AFM current images were obtained with a Bruker Icon atomic force microscope. Contact angle was tested on an automatic contact angle tester (JY-82A). The photo-CELIV curves were measured with a commercially available Paios system (FLUXiM AG), and an average mobility was calculated based on the equation of ({mu }_{{{mathrm{avg}}}}={2d}^{2}/(3A{t}_{max }^{2}(1+0.36Delta j/{j}_{0}))), where d is active layer thickness, A is voltage rise speed of the applied voltage pulse, tmax is the time to reach the extraction current maximum, Δj is the displacement current, and j0 is initial current. UV-vis absorption spectra were obtained with a Metash UV-8000 spectrometer. UPS and XPS spectra were acquired with Thermo Fisher Scientific XPS Escalab Xi+. Both steady-state PL and TRPL spectra were acquired on a HORIBA FL-3 fluorescence spectrophotometer at room temperature. The perovskite films were coated onto HTL-covered glass substrates. The excitation wavelengths were 480 and 405 nm for PL and TRPL measurements, respectively. In TPV measurements, the devices were placed under background light bias enabled by a focused Quartz Tungsten-Halogen Lamp with an intensity of 100 mW cm−2. Photo-excitations were generated with 8 ns pulses from a laser system (Oriental Spectra, NLD520). A digital oscilloscope was used to acquire the TPV signal at the open-circuit condition. TPC signals were measured under short-circuit conditions under the same excitation wavelength without background light bias. 1H NMR and 13C NMR spectra were obtained on a 400 MHz Bruker AVANCE III-400 spectrometer. tDOS and DLCP spectra were obtained with an Agilent E4980A precision LC meter.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
All data generated or analyzed in this study are provided in the published article and its Supplementary Information. Source data are provided as a Source data file. Source data are provided with this paper.
Zhang, X. et al. Advances in inverted perovskite solar cells. Nat. Photonics 18, 1243–1253 (2024).
Article ADS CAS Google Scholar
Feng, K. et al. Non-fullerene electron-transporting materials for high-performance and stable perovskite solar cells. Nat. Mater. 24, 770–777 (2025).
Article ADS CAS PubMed Google Scholar
Shen, Z. et al. Precise synthesis of advanced polyarylamines for efficient perovskite solar cells. Nat. Mater. 24, 1450–1456 (2025).
Article ADS CAS PubMed Google Scholar
Luo, X. et al. Recent Advances of Inverted Perovskite Solar Cells. ACS Energy Lett. 9, 1487–1506 (2024).
Article CAS Google Scholar
Wang, F.-F. et al. Simplified p-i-n Perovskite Solar Cells with a Multifunctional Polyfullerene Electron Transporter. Chin. J. Polym. Sci. 42, 1060–1066 (2024).
Article CAS Google Scholar
He, R. et al. Improving interface quality for 1-cm2 all-perovskite tandem solar cells. Nature 618, 80–86 (2023).
Article ADS CAS PubMed Google Scholar
Azmi, R. et al. Double-side 2D/3D heterojunctions for inverted perovskite solar cells. Nature 628, 93–98 (2024).
Article ADS CAS PubMed Google Scholar
Tang, H. et al. Reinforcing self-assembly of hole transport molecules for stable inverted perovskite solar cells. Science 383, 1236–1240 (2024).
Article ADS CAS PubMed Google Scholar
Dong, B. et al. Self-assembled bilayer for perovskite solar cells with improved tolerance against thermal stresses. Nat. Energy 10, 342–353 (2025).
Article ADS CAS Google Scholar
Qu, G. et al. Self-assembled materials with an ordered hydrophilic bilayer for high performance inverted Perovskite solar cells. Nat. Commun. 16, 86 (2025).
Article ADS CAS PubMed PubMed Central Google Scholar
Li, M., Liu, M., Qi, F., Lin, F. R. & Jen, A. K. Y. Self-assembled monolayers for interfacial engineering in solution-processed thin-film electronic devices: design, fabrication, and applications. Chem. Rev. 124, 2138–2204 (2024).
Article CAS PubMed Google Scholar
Li, B. et al. Fundamental understanding of stability for halide perovskite photovoltaics: The importance of interfaces. Chem. 10, 35–47 (2024).
Article CAS Google Scholar
Boyd, C. C., Cheacharoen, R., Leijtens, T. & McGehee, M. D. Understanding degradation mechanisms and improving stability of perovskite photovoltaics. Chem. Rev. 119, 3418–3451 (2019).
Article CAS PubMed Google Scholar
Zhu, H. et al. Long-term operating stability in perovskite photovoltaics. Nat. Rev. Mater. 8, 569–586 (2023).
Article ADS Google Scholar
Thiesbrummel, J. et al. Ion-induced field screening as a dominant factor in perovskite solar cell operational stability. Nat. Energy 9, 664–676 (2024).
Article ADS CAS Google Scholar
Liang, Q. et al. Highly stable perovskite solar cells with 0.30 voltage deficit enabled by a multi-functional asynchronous cross-linking. Nat. Commun. 16, 190 (2025).
Article ADS PubMed PubMed Central Google Scholar
Gao, Z.-W. et al. Eutectic molecule ligand stabilizes photoactive black phase perovskite. Nat. Photonics 19, 258–263 (2025).
Article ADS CAS Google Scholar
Xing, Z. et al. Solubilizing and stabilizing C60 with n-type polymer enables efficient inverted perovskite solar cells. Joule 9, 101817 (2025).
Article CAS Google Scholar
Leijtens, T. et al. Stability of metal halide perovskite solar cells. Adv. Energy Mater. 5, 1500963 (2015).
Article Google Scholar
Razera, R. A. Z. et al. Instability of p–i–n perovskite solar cells under reverse bias. J. Mater. Chem. A 8, 242–250 (2020).
Article ADS CAS Google Scholar
Guo, X. et al. Mitigating surface deficiencies of perovskite single crystals enables efficient solar cells with enhanced moisture and reverse-bias stability. Adv. Funct. Mater. 33, 2213995 (2023).
Article CAS Google Scholar
Aeberhard, U. et al. Multi-scale simulation of reverse-bias breakdown in all-perovskite tandem photovoltaic modules under partial shading conditions. Sol. RRL 8, 2400492 (2024).
Article Google Scholar
Aydin, E. Raising the bar for breakdown. Nat. Energy 9, 1183–1184 (2024).
Article ADS Google Scholar
Wang, C. et al. Perovskite solar cells in the shadow: understanding the mechanism of reverse-bias behavior toward suppressed reverse-bias breakdown and reverse-bias induced degradation. Adv. Energy Mater. 13, 2203596 (2023).
Article CAS Google Scholar
Bowring, A. R., Bertoluzzi, L., O’Regan, B. C. & McGehee, M. D. Reverse bias behavior of halide perovskite solar cells. Adv. Energy Mater. 2018, 1702365 (2017).
Google Scholar
Jiang, F. et al. Improved reverse bias stability in p–i–n perovskite solar cells with optimized hole transport materials and less reactive electrodes. Nat. Energy 9, 1275–1284 (2024).
Article ADS CAS Google Scholar
Li, N. et al. Barrier reinforcement for enhanced perovskite solar cell stability under reverse bias. Nat. Energy 9, 1264–1274 (2024).
Article ADS CAS Google Scholar
Wu, W. et al. Bilateral alkylamine for suppressing charge recombination and improving stability in blade-coated perovskite solar cells. Sci. Adv. 5, eaav8925 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Isikgor, F. H. et al. Scaling-up perovskite solar cells on hydrophobic surfaces. Nano Energy 81, 105633 (2021).
Article CAS Google Scholar
Kuan, C.-H. et al. How can a hydrophobic polymer PTAA serve as a hole-transport layer for an inverted tin perovskite solar cell?. Chem. Eng. J. 450, 138037 (2022).
Article CAS Google Scholar
Wang, L. et al. Electrophilic molecule-induced π–π interactions reduce energy disorder of the hole transport layer for highly efficient perovskite solar modules. Energy Environ. Sci. 17, 8337–8348 (2024).
Article CAS Google Scholar
Zhang, W. et al. Synergistic effects of the physical modification and chemical passivation enabling efficient perovskite solar cells. Chem. Eng. J. 497, 154864 (2024).
Article CAS Google Scholar
Ren, Z. et al. Poly(carbazole phosphonic acid) as a versatile hole-transporting material for p-i-n perovskite solar cells and modules. Joule 7, 2894–2904 (2023).
Article CAS Google Scholar
Wang, F. et al. A polymeric hole transporter with dual-interfacial interactions enables 25%-efficiency blade-coated perovskite solar cells. Adv. Mater. 36, 2412059 (2024).
Article CAS Google Scholar
Zhao, Y. et al. Enhanced interface adhesion with a polymeric hole transporter enabling high-performance air-processed perovskite solar cells. Energy Environ. Sci. 18, 1366–1374 (2025).
Article CAS Google Scholar
Zhang, S. et al. Conjugated self-assembled monolayer as stable hole-selective contact for inverted perovskite solar cells. ACS Mater. Lett. 4, 1976–1983 (2022).
Article CAS Google Scholar
Jiang, C. et al. Double layer composite electrode strategy for efficient perovskite solar cells with excellent reverse-bias stability. Nano-Micro Lett. 15, 12 (2022).
Article ADS CAS Google Scholar
Zheng, X. et al. Co-deposition of hole-selective contact and absorber for improving the processability of perovskite solar cells. Nat. Energy 8, 462–472 (2023).
Article ADS CAS Google Scholar
Duan, Y. et al. A comprehensive review of organic hole-transporting materials for highly efficient and stable inverted perovskite solar cells. Adv. Funct. Mater. 34, 2315604 (2024).
Article CAS Google Scholar
El-Hajje, G. et al. Quantification of spatial inhomogeneity in perovskite solar cells by hyperspectral luminescence imaging. Energy Environ. Sci. 9, 2286–2294 (2016).
Article CAS Google Scholar
Tegetmeyer, B. et al. Electroluminescence from silicon vacancy centers in diamond p–i–n diodes. Diam. Relat. Mater. 65, 42–46 (2016).
Article ADS CAS Google Scholar
Wang, C. et al. Abnormal dynamic reverse bias behavior and variable reverse breakdown voltage of ETL-free perovskite solar cells. Sol. RRL 7, 2300456 (2023).
Article CAS Google Scholar
Sarritzu, V. et al. Optical determination of Shockley-Read-Hall and interface recombination currents in hybrid perovskites. Sci. Rep. 7, 44629 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Shao, Y., Xiao, Z., Bi, C., Yuan, Y. & Huang, J. Origin and elimination of photocurrent hysteresis by fullerene passivation in CH3NH3PbI3 planar heterojunction solar cells. Nat. Commun. 5, 5784 (2014).
Article ADS CAS PubMed Google Scholar
Ni, Z. et al. Resolving spatial and energetic distributions of trap states in metal halide perovskite solar cells. Science 367, 1352–1358 (2020).
Article ADS CAS PubMed Google Scholar
Zhao, J. et al. Is Cu a stable electrode material in hybrid perovskite solar cells for a 30-year lifetime? Energy Environ. Sci. 9, 3650–3656 (2016).
Article CAS Google Scholar
Lanzetta, L. et al. Tin–lead perovskite solar cells with enhanced reverse bias stability. ACS Energy Lett. 10, 2093–2095 (2025).
Article CAS Google Scholar
Kaltenbrunner, M. et al. Flexible high power-per-weight perovskite solar cells with chromium oxide–metal contacts for improved stability in air. Nat. Mater. 14, 1032–1039 (2015).
Article ADS CAS PubMed Google Scholar
Guerrero, A. et al. Interfacial degradation of planar lead halide perovskite solar cells. ACS Nano 10, 218–224 (2015).
Article PubMed Google Scholar
Hu, S. et al. Steering perovskite precursor solutions for multijunction photovoltaics. Nature 639, 93–101 (2025).
Article ADS CAS PubMed Google Scholar
Download references
This work was supported by the National Key Research and Development Program of China (No. 2023YFB3809700), the National Natural Science Foundation of China (No. 52372196), and Jiangsu Double Innovation Team Program, and the open research fund of Suzhou Laboratory (SZLAB-1308-2024-TS007).
These authors contributed equally: Chaoyue Zhao, Feifei Wang, Xiaodong Hu.
State Key Laboratory of Coordination Chemistry, MOE Key Laboratory of High-Performance Polymer Materials & Technology, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing, China
Chaoyue Zhao, Feifei Wang, Xiaodong Hu, Yaoyao Zhang, Tianxiao Liu, Yangyang Liu, Lingyuan Wang, Xiaoyu Shi, Xinsheng Tang, Wei Wang & Shangshang Chen
Department of Chemistry and Hong Kong Branch of Chinese National Engineering Research Center for Tissue Restoration and Reconstruction, Hong Kong University of Science and Technology, Hong Kong, China
Chaoyue Zhao, Siwei Luo & He Yan
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
S.C. conceived the idea for the study and designed the experiments. C.Z. fabricated and characterized PSCs. F.W. synthesized Poly-PhPACz. X.H. assisted with the device fabrication. Y.Z. carried out EL mapping. T.L. and Y.L. assisted with device characterization. L.W. conducted c-AFM characterization. X.S. tested the device stability. S.L. and X.T. assisted with film analysis and device characterization. H.Y. supervised the device fabrication. W.W. supervised EL mapping. C.Z. and S.C. wrote the manuscript with inputs from all co-authors. All authors discussed the results and reviewed the manuscript. C.Z., F.W., and X.H. contributed equally to this work.
Correspondence to Shangshang Chen.
The authors declare no competing interests.
Nature Communications thanks Cong Liu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
Zhao, C., Wang, F., Hu, X. et al. Ambient blade-coated perovskite solar cells with high reverse bias stability enabled by polymeric hole transporter design. Nat Commun 16, 10355 (2025). https://doi.org/10.1038/s41467-025-65341-7
Download citation
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41467-025-65341-7
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
Nano-Micro Letters (2026)
Journal of Materials Science: Materials in Energy (2026)
Advertisement
Nature Communications (Nat Commun)
ISSN 2041-1723 (online)
© 2026 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
