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 17, Article number: 711 (2026) Cite this article
7697 Accesses
3 Citations
Metrics details
Monolithic perovskite/Cu(In,Ga)Se₂ (CIGS) tandem solar cells offer unique advantages among tandem configurations, including optimal bandgap pairing, all-thin-film architecture, superior radiation hardness, and exceptional stability. Despite their potential, challenges remain in optimizing efficiency and stability, particularly in the intermediate recombination layer (IRL) that connects subcells. This study addresses these challenges by developing a high-performance IRL using a composite structure comprising Al:ZnO/Au/NiO_x/[4-(7H-dibenzo[c,g]carbazol-7-yl)butyl] phosphonic acid (4PADCB). The hybrid NiO_x/4PADCB hole transport layer enhances interface trap passivation, optimizes band alignment, and improves minority carrier extraction, while the ultrathin Au layer significantly boosts the majority carrier recombination rate. With well-engineered subcells, our champion perovskite/CIGS tandem device achieved a record PCE of 28.04% at 0.51 cm² area and 30.71% at 0.15 cm² area (30.1% cross-verified by an external organization), with an exceptional fill factor of 80.9%, alongside outstanding photo- and thermal-stability. This work establishes monolithic perovskite/CIGS tandems as competitive with leading perovskite/Si (34.9%) and perovskite/perovskite (30.1%) technologies, providing a scalable, versatile framework for next-generation photovoltaics.
Perovskite-based tandem solar cells are emerging as a leading photovoltaic technology due to their potential for achieving high power conversion efficiency (PCE) at low production costs. Among the various options for narrow-bandgap bottom subcells, including silicon^1,2,3,4,5, CIGS^{6,7,8,9,10,11}, and low-bandgap perovskite^12,13,14,15, the perovskite/Si tandem is the most extensively studied and has achieved a record PCE of 34.9%^16,17. In contrast, the perovskite/CIGS tandems present unique advantages over their counterparts. With tunable CIGS band gaps as low as 1.00 eV, they can provide a bandgap pairing closest to the theoretical requirement: 1.62 eV for the top cell and 0.96 eV for the bottom cell to obtain the maximum theoretical efficiency¹⁸. As an all-thin-film technology, they possess a device architecture compatible with flexible and lightweight photovoltaic applications⁶. Moreover, the radiation hardness of both subcells makes them particularly well-suited for space applications^19,20.
A number of studies have focused on enhancing the performance of two-terminal monolithic perovskite/CIGS tandems, particularly addressing the intermediate recombination layer (IRL) that connects the two subcells^6,9,11,21. Early work employed a chemical-mechanical polished indium tin oxide (ITO) layer to mitigate the rough surface of the CIGS subcell, achieving a certified PCE of 22.4%⁶. To avoid surface damages introduced by polishing, subsequent work applied conformal thin NiO_x layers or self-assembled monolayers (SAMs), such as [2-(3,6-dimethoxy-9H-carbazol-9-yl)ethyl]phosphonic acid (MeO-2PACz) and [4-(3,6-dimethyl-9H-carbazol-9-yl)butyl]phosphonic acid (Me-4PACz), directly on the as-grown transparent conducting oxide surface of CIGS subcells. These approaches achieved certified PCEs of 21.6%⁹, 23.3%¹⁰, and 24.2%¹¹ on rigid substrates. In the meantime, notable progress has also been made on flexible perovskite/CIGS tandem solar cells, reaching efficiencies of 23.64% and 23.28%^22,23. The most recent advance used a NiO_x/[2-(9H-carbazol-9-yl)ethyl]phosphonic acid (2PACz) double layer on a relatively smooth CIS subcell, yielding a record PCE of 24.9% (certified 23.5%)²¹ on rigid substrates, and a NiO_x/4PADCB configuration enabled a flexible perovskite/CIGS tandem to reach 22.8%²⁴, nearly concurrent with this work. Despite these achievements, the progress of monolithic perovskite/CIGS tandems remained far behind perovskite/Si tandems (34.9%) and perovskite/perovskite tandems (30.1%)^16,17. Notably, over the course of nearly 8 years, the efficiency of monolithic perovskite/CIGS tandems improved slightly from 22.4%⁶ (2018), 24.2%¹¹ (2022), to 26.3%¹⁶ (2025), reflecting the tremendous challenges that remain in pushing this technology forward. These challenges primarily stem from non-optimized subcell efficiencies, mismatched band gaps, and, critically, the lack of efficient IRLs (Supplementary Table 1).
To address these limitations, we developed a composite IRL consisting of Al:ZnO(AZO)/Au/NiO_x/4PADCB for monolithic perovskite/CIGS tandems. The multi-layer HTL was designed to leverage the suitable work function and optoelectronic properties of NiO_x²⁵, which conform well to the rough surface of the CIGS bottom subcell to suppress microscale shunt paths. The addition of 4PADCB not only passivates NiO_x surface defects, reduces interfacial recombination, and enhances hole extraction through favorable energy level alignment, but also facilitates high-quality perovskite film formation owing to its excellent solubility, wetting behavior, and large dipole moment compared to other SAMs.^15,26,27 An ultrathin Au layer is further introduced between the subcells to promote efficient majority carrier recombination. Notably, this composite IRL can be directly applied to CIGS subcells without requiring surface polishing. As a result, we achieved perovskite/CIGS tandems with high open-circuit voltage (V_OC, 1.74 V), substantial short-circuit current density (J_SC, 21.76 mA/cm²), and an exceptional fill factor (FF, 80.90%). The champion device, with an active area of 0.15 cm² and featuring a 1.01 eV bandgap bottom subcell and a 1.67 eV bandgap top subcell, achieved a PCE of 30.71% (30.10% cross-verified by an external organization). This represents a remarkable absolute increase of nearly 4% improvement over the previous certified record of 26.3%¹⁶, making it another 2-terminal tandem with efficiency exceeding 30%. Our achievement places perovskite/CIGS tandems alongside leading technologies like perovskite/Si (34.9%) and perovskite/perovskite (30.1%), as shown in the NREL chart¹⁶. Furthermore, the composite IRL suppresses photo-induced segregation in the perovskite layer and enhances the stability of unencapsulated tandem devices. Overall, this design not only significantly improves efficiency and stability but also offers a versatile framework that can be adapted to other tandem systems, broadening its potential impact on high-efficiency photovoltaic technologies.
The advantages of using a NiO_x/4PADCB double layer as a hybrid HTL for the perovskite top subcell were clearly demonstrated in Fig. 1. Through careful optimization of the CIGS fabrication process (Supplementary Figs. 1–8, Table 2), we successfully deposited the HTL directly onto the AZO transparent electrode surface of the as-grown CIGS subcell. X-ray photoelectron spectroscopy (XPS) results also confirmed that all three hole transport materials could be effectively deposited onto the sputtered NiO_x surface (Supplementary Fig. 9). When 4PADCB was deposited alone onto the AZO surface (4PADCB), the devices exhibited a very low FF (~35%) and a corresponding low PCE (~ 12%). Using sputtered NiO_x alone (Sput) significantly improved the FF to ~68%, as it effectively covered the surface of the CIGS subcell and eliminated microscale shunt paths. However, defects on the NiO_x surface limited the V_OC to a low level of ~1.57 V. By combining sputtered NiO_x with 4PADCB (Sput/4PADCB), surface defects were passivated, improving the FF to ~71% and boosting the V_OC to ~1.71 V. Replacing the sputtered NiO_x with spin-coated NiO_x (Spin/4PADCB) reduced both FF (~ 56%) and V_OC (~ 1.59 V), highlighting the superior ability of sputtered NiO_x to eliminate the leakage between the top and bottom subcells. As discussed in the introduction and detailed in Supplementary Table 1, the various IRLs used in prior perovskite/CIGS tandem studies have consistently suffered from either significant V_OC losses or low FF, or both^6,9,11,21. These limitations often stemmed from incomplete interfacial defect passivation and inadequate shunt path removal. In contrast, our hybrid HTL with 4PADCB SAM on sputtered NiO_x successfully addressed these issues.
a Schematic of the tandem solar cell structure incorporating different HTLs. b–e Comparison of key performance parameters: (b) V_OC, (c) FF, (d) J_SC, and (e) PCE for tandem cells with various HTLs indicated on the horizontal axis. Each data point corresponds to a statistical set (n ≥ 18). The open circle, top and bottom whiskers, and box indicate the mean, minimum and maximum values, and the 25%–75% interquartile range, respectively.
To maximize the benefits of the hybrid HTL, we carefully controlled the thickness and conductance of the sputtered NiO_x layer. As shown in Supplementary Fig. 10, the optimal thickness was determined to be ~20 nm. A thinner NiO_x layer failed to adequately cover the rough surface of CIGS subcells, leading to incomplete shunt path elimination and reduced FF and V_OC. Conversely, a thicker NiO_x layer introduced increased internal resistance and carrier recombination, which also negatively impacted V_OC. Conductivity of the NiO_x layer also played a critical role. As demonstrated in Supplementary Fig. 11, Table 3, incorporating oxygen into the sputtering atmosphere increased the Ni³⁺ density and the conductivity²⁸, resulting in improved FF and V_OC. However, excessive interstitial oxygen and Ni³⁺ ions would reduce the film quality and decrease FF. The optimal O₂/(Ar+O₂) pressure ratio was therefore established at 1%. As shown in Supplementary Fig. 12, both Grazing Incidence X-ray Diffraction (GI-XRD) and Raman analyzes indicate the formation of pure-phase nickel oxide phase without metallic nickel, with structural and conductivity changes correlating to oxygen content.
To evaluate the performance of 4PADCB against other organic hole transport materials, we combined sputtered NiO_x with two commonly used materials in single-junction perovskite solar cells: 2PACz and poly(N, N’-bis-4butylphenyl-N, N’-bisphenyl)benzidine (TPD)^29,30,31 (Supplementary Fig. 13). As illustrated in Fig. 1, the devices exhibited a V_OC of ~1.69 V and an FF of ~67% for Sput/2PACz and ~1.66 V and ~63% for Sput/TPD. With comparable J_SC across all devices, the average PCEs incorporating TPD, 2PACz, and 4PADCB yielded 22.0%, 23.5%, and 25.5%, respectively, affirming 4PADCB as the superior choice for the hybrid HTLs. To understand these performance differences, both film thickness and molecular properties of the HTLs were considered. TPD usually forms a relatively thicker layer, potentially increasing internal resistance and reducing FF. In contrast, 4PADCB and 2PACz, as self-assembled monolayers, are much thinner and thus exert minimal influence on resistance. Beyond thickness, the advantages of 4PADCB are also linked to its distinct molecular structure, stronger interfacial dipole, and denser monolayer packing^15,32,33.
Furthermore, we found that employing a thicker perovskite layer than is typically used in single-junction devices appreciably improved the performance of perovskite/CIGS tandems. In contrast to the double-pass light absorption that occurred in single-junction perovskite solar cells enabled by metallic electrode reflection, top subcells in tandems relied solely on single-pass absorption. As shown in Supplementary Figs. 14, 15, increasing the concentration of perovskite precursor solution would increase the average layer thickness, raising J_SC from ~20.6 mA/cm² at 1.0 mmol/mL to ~21.6 mA/cm² at 2.5 mmol/mL due to enhanced light absorption. Moreover, the thicker perovskite layer buried AZO protrusions more effectively, further mitigating microscale shunt paths and improving both V_OC and FF. Nevertheless, excessively increasing the perovskite thickness introduced tremendous non-uniformity, increasing internal resistance, carrier transport path, and recombination rate while limiting the isolation of AZO protrusions. These effects adversely affected J_SC, V_OC, and FF. Balancing these factors, an optimal perovskite thickness of ~900 nm was determined (Supplementary Fig. 14), which aligns well with findings for perovskite/Si tandems.
To uncover the mechanisms behind the superior performance of the Sput/4PADCB hybrid HTL, we investigated the interface quality and energy level alignment using advanced characterization techniques. Confocal fluorescence lifetime imaging microscopy (FLIM) was employed to analyze the spatial distribution of photoluminescence (PL) lifetimes for perovskite films deposited on different HTLs. To eliminate interference from the rough CIGS surface, we used a smooth ITO substrate to form an ITO/HTL/perovskite stack. As shown in Fig. 2a, c, d, samples with 4PADCB, Spin/4PADCB, and Sput/TPD all showed relatively short PL lifetimes and considerable inhomogeneity. In contrast, samples Sput, Sput/2PACz, and Sput/4PADCB demonstrated enhanced PL lifetimes and notably improved uniformity (Fig. 2 b, e, f), with Sput/4PADCB performing the best. These results were further corroborated by surface potential measurements using Kelvin probe force microscopy (KPFM) (Supplementary Figs. 16, 17, Table 4) on different HTLs deposited on CIGS bottom cells under both ambient air and nitrogen environments. Large-area time-resolved PL (TRPL) measurements further quantified the improvement, with average carrier lifetimes of 15, 31, 74, 107, 161, and 217 ns, respectively, for Spin/4PADCB, 4PADCB, Sput/TPD, Sput, Sput/2PACz, and Sput/4PADCB (Supplementary Fig. 18, Table 5). Consistent trends were observed in TRPL measurements with excitation from both sides of the samples, as shown in Supplementary Fig. 19. The trend in steady-state PL intensity aligned with these findings, where higher PL intensities corresponded to longer carrier lifetimes. The Sput/4PADCB hybrid HTL demonstrated the most uniform interface, the longest radiative lifetime, and the highest PL intensity, confirming its superior ability to suppress non-radiative recombination at the perovskite/HTL interface.
a–f Photoluminescence lifetime images for perovskite films deposited on various HTLs: (a) 4PADCB, (b) Sput, (c) Spin/4PADCB, (d) Sput/TPD, (e) Sput/2PACz, and (f) Sput/4PADCB. g Energy level diagram illustrating the alignment between various HTLs and a wide-bandgap perovskite film. h Transient surface photovoltage for half-stack configurations of ITO/HTL/Perovskite/Au-Electrode with various HTLs.
To assess the hole extraction efficiency, we used ultraviolet photoelectron spectroscopy (UPS) to examine energy level alignment and transient surface photovoltage (tr-SPV) measurement to study the hole transport dynamics. As depicted in Supplementary Fig. 20, UPS data revealed the work functions and energy positions of the highest occupied molecular orbital (HOMO) or valence band maximum (VBM) for the various HTLs and perovskite layers. All HTLs displayed VBMs/HOMOs between 680 and 750 meV below the Fermi level, which were higher than that of the perovskite, enabling energetically possible hole extraction. As further demonstrated in Fig. 2g, in addition to a large valence band offset (~600, ~630, ~790 meV, respectively) for spin-coated NiO_x, sputtered NiO_x, and TPD-coated NiO_x, the work function differences introduced an interface electric field between HTL and perovskite, resulting in a high energy barrier of ~300–520 meV. The large band offsets and energy barriers hindered hole transport from the perovskite layers to the HTLs, limiting extraction efficiency. In contrast, when 2PACz and 4PADCB were deposited on sputtered NiO_x, their strong molecular dipole moment increased the work functions of the hybrid HTLs and reversed the interface electric field direction. The resulting small valence band offsets (~250 meV for 2PACz and ~190 meV for 4PADCB) and favorable electric fields facilitated efficient hole extraction.
The effectiveness of hole extraction for these interfaces was further validated by tr-SPV measurements. As shown in Fig. 2h, the tr-SPV signal from the ITO/HTL/perovskite/Au-Electrode half-stacks demonstrated about three-fold increase in amplitude, from −1.7 mV to −5.3 mV, accompanied by a reduction in extraction time from 53.3 ns to 36.0 ns in the sequence of Sput, Sput/TPD, Sput/2PACz, Sput/4PADCB (Supplementary Fig. 21a, Table 6). Similar trends were observed under nitrogen environments, with extraction times reducing from 67.3 ns to 23.3 ns (Supplementary Fig. 21b, Table 7). These results and the observed steep voltage rise for Sput/4PADCB indicated that the perovskite/4PADCB/NiO_x junction enabled the fastest and most efficient transport of photo-generated holes among all configurations tested. Collectively, the data indicated that 4PADCB provides multiple benefits when combined with sputtered NiO_x: it passivates interfacial defects, minimizing carrier trapping and extending PL lifetimes, and it significantly improves energy level alignment, reducing transport barriers and enhancing hole extraction rates. These improvements translate directly to higher V_OC and FF for the tandem devices with sputtered NiO_x/4PADCB hybrid HTL.
The proper IRL in a monolithic p-i-n (bottom)/p-i-n (top) tandem must consist of an electron transport layer (ETL) and an HTL that not only efficiently extracts electrons and holes from their respective light absorbers, but also recombines these carriers effectively and timely at the ETL/HTL junction³⁴. For the optimized AZO/NiO_x/4PADCB IRL, we introduced an ultrathin Au particle layer (Supplementary Fig. 22) at the AZO/NiO_x interface to further improve device performance. As shown in Fig. 3a–d, when the average Au thickness increased from 0 to 0.6 nm, the J_SC exhibited only a slight fluctuation, remaining around ~21.3 to ~21.2 mA/cm², possibly due to minor light absorption by the Au particles. More importantly, the FF improved significantly from ~71% to ~76%. With the simultaneously increased V_OC from ~1.71 V to ~1.73 V, the average PCE boosted from 25.6% to 27.6%. When the average Au thickness exceeded 1 nm, FF saturated, and the overall device performance declined. These results highlight the critical role of the ultrathin Au particle layer in the AZO/Au/NiO_x/4PADCB structure for enhancing FF and overall device performance.
a–d Comparison of V_OC (a), FF (b), J_SC (c), and PCE (d) for tandem devices with varying average Au layer thickness. e Schematic of the energy level diagram of the AZO/Au/NiO_x/4PADCB recombination junction. f J–V curves for the champion tandem device. g External quantum efficiency (EQE) of the tandem device, where the slightly lower integrated J_SC may result from insufficient light intensity to saturate metastable defects. h Historical progress of record efficiencies for monolithic perovskite/CIGS tandem devices. Each data point corresponds to a statistical set (n ≥ 22). The open circle, top and bottom whiskers, and box indicate the mean, minimum and maximum values, and the 25%–75% interquartile range, respectively.
To understand the role played by the thin Au particle layer, we plotted the energy level diagram of the AZO/Au/NiO_x/4PADCB/perovskite stack in Fig. 3e. As shown in Supplementary Fig. 20, the VBM of sputtered NiO_x is close to the HOMO of 4PADCB and approximately 700 meV below the conduction band minimum (CBM) of the heavily doped AZO. Thus, the contribution from direct band-to-band tunneling to the carrier recombination between electrons from AZO and holes from NiO_x is negligible. This conclusion is supported by the resistive-like current-voltage (J–V) characteristics observed in the AZO/NiO_x/4PADCB and AZO/Au/NiO_x/4PADCB junctions (Supplementary Fig. 23). Without the Au particle layer, the carrier recombination was primarily mediated by interfacial trap states. Introducing the ultrathin Au particle layer opened up an additional recombination channel through Au particle states, and the insertion of an ultrathin Au layer could induce band bending at the Au/NiO_x interface due to the relatively high work function of Au, thereby facilitating hole extraction and increasing the recombination rate for majority carriers by ~12%, as evidenced by the increased current density of ~12% (Supplementary Fig. 23). This timely recombination of electrons and holes in IRL eliminated the accumulation of un-recombined charge carriers at the junction, reducing the reverse built-in electric field. The reduced charge carrier accumulation within the IRL due to the incorporation of the Au layer led to more efficient extraction of holes from the perovskite/4PADCB/NiO_x interface, thus reducing the residence time of holes and consequently the number of minority carrier recombination. A similar argument applies to electrons at the CIGS/CdS interface. In the end, a higher FF and V_OC for the tandem devices could be achieved, as evidenced by the improved J–V and dark J–V characteristics in Supplementary Fig. 24.
In addition to the Au particle layer, the AZO ETL layer was optimized to complete the ideal IRL. As shown in Supplementary Fig. 25, the optimal AZO thickness was determined to be ~200 nm. Thinner AZO significantly reduced both FF and V_OC, likely due to ineffective electron collection, while thicker layers decreased J_SC, likely due to parasitic absorption.
For the present tandem structure, incident light had to pass the C₆₀ layer before reaching the perovskite layer. The absorptive nature of C₆₀ in the ultraviolet (UV) region of the solar spectrum could also affect the performance of the top subcell. As demonstrated in Supplementary Figs. 26, 27 for single semi-transparent perovskite solar cells, a normal thickness of C₆₀ at 20 nm indeed reduced the J_SC by incidence light from the top TCO side, and a reduced thickness at ~12 nm could reach a better overall performance. As further illustrated in Supplementary Fig. 28, an optimal thickness of ~12 nm, in contrast to ~20 nm normally used for stand-alone opaque perovskite solar cells, could effectively avoid strong UV parasitic light absorption and maintain reasonable V_OC and FF for the tandem devices.
Using the optimized composite IRL of AZO/Au/NiO_x/4PADCB, along with previously optimized bottom and top subcells (Supplementary Fig. 29)³⁵, we fabricated a champion cell with a 0.15 cm² active area. The device achieved a PCE of 30.71%, with a V_OC of ~1.745 V, an FF of ~80.9%, and a J_SC of ~21.76 mA/cm² (Fig. 3f, corresponding EQE spectra shown in Fig. 3g). The PCE cross-verified by an external organization was 30.1%, with a V_OC ~ 1.731 V, an FF ~ 80.2%, and a J_SC ~ 21.67 mA/cm² (Supplementary Note 1). Additionally, a larger area of 0.51 cm² tandem device was fabricated, achieving an efficiency of 28.04% (Supplementary Fig. 30). As shown in Fig. 3h, this work marks a significant milestone in the development of monolithic perovskite/CIGS tandem solar cells. Our device not only sets a notably high efficiency for monolithic perovskite/CIGS tandems but also surpasses the performance of four-terminal architectures (Supplementary Table 8). Furthermore, it also exceeds the record efficiencies of both constituent single-junction solar cells in their optimized configurations (perovskite: 27.0%, CIGS: 23.6%, NREL chart, 2025)¹⁶. Compared to the previous champion perovskite/CIGS tandem devices (Supplementary Table 1), the significantly improved PCE can be attributed primarily to the remarkable enhancement in both FF and J_SC. The moderate V_OC resulted from deliberate bandgap optimization, with the CIGS bottom subcell set at 1.01 eV and the perovskite top subcell at 1.67 eV, enabling efficient current matching. The exceptional FF value exceeding 80%, one of the highest reported for perovskite/CIGS tandems²², surpassed that of standalone CIGS subcells, demonstrating the effectiveness of the optimized IRL. The high J_SC was achieved through the use of a narrow-bandgap CIGS bottom subcell, a thicker perovskite absorber, optimized bandgap matching, and an anti-reflection coating.
While 4-terminal (4 T) perovskite/CIGS tandems have shown high efficiency^7,23 and favorable energy yield under real-world conditions, detailed modeling has demonstrated that 4 T configurations deliver only a modest increase in annual energy yield compared to their 2-terminal (2 T) counterparts, typically about 1–3%, depending on deployment location and tracking strategy³⁶. In earlier studies, the monolithic 2 T devices have so far exhibited lower performance, their limitations primarily stem from non-optimized interfacial recombination layers rather than intrinsic material or structural constraints. Here, we demonstrate a monolithic perovskite/CIGS tandem with the PCE exceeding 30% through the introduction of a multifunctional AZO/Au/NiO_x/4PADCB interlayer, highlighting its potential for scalable, high-yield photovoltaic applications.
We evaluated the photostability of perovskite films with various hole transport layers in ambient conditions under continuous-wave laser illumination at different light intensities. Time-dependent and excitation light intensity-dependent PL measurements were conducted at 405 nm wavelength. As shown in Fig. 4a, at 1-sun-equivalent intensity, all five perovskite films exhibited minimal changes in their PL spectra, indicating good photostability under standard operating conditions. To accelerate or amplify the light-induced phase segregation process, the illumination intensity was further increased to 10- and 30-sun equivalents, aiming to elucidate how different interfacial structures affect device stability. Notably, only the Spin/4PADCB and Sput/4PADCB samples maintained nearly unchanged PL spectra even under such high-intensity illumination. Supplementary Fig. 31 presents the PL spectra normalized to the maximum value within each data set (i.e., from 0 to 20 min), which facilitates the observation of intensity changes over time. Considering both PL intensity and spectral shifts, 4PADCB appears to exhibit relatively better stability compared to 2PACz and TPD. Light-induced halide segregation in wide-bandgap perovskites with high Br/I ratios is often linked to PL instability due to the formation of iodide-rich clusters. While charge accumulation has been proposed to accelerate this process by promoting ion migration^32,37,38, the segregation mechanism is likely multifactorial and still requires further clarification.
a Normalized evolution of PL spectra for perovskite films on different HTLs (4PADCB, Spin/4PADCB, Sput/TPD, Sput/2PACz, and Sput/4PADCB) under equivalent 1-sun, 10-sun, and 30-sun illumination over 20 min. b Long-term storage stability of monolithic perovskite/CIGS tandem devices at room temperature (25 °C) in an N₂ atmosphere. c Thermal stability of tandem devices at 65 °C in an N₂ atmosphere. d Continuous MPPT test for unencapsulated tandem devices. Note that absolute efficiency data for Figures (b–d) are provided in Supplementary Fig. 33, and these measurements were all performed under standard 1-sun illumination conditions.
Due to their ultrathin nature, SAMs can rapidly transport photo-generated holes, effectively suppressing light-induced phase segregation^10,32,38. Although 4PADCB and 2PACz share identical anchoring groups and exhibit similar defect passivation on sputtered NiO_x surfaces, they differ in their functional groups: 4PADCB possesses a 7H-dibenzo[c,g]carbazole, while 2PACz contains a carbazole. The functional group in 4PADCB features expanded π-orbitals arranged in a non-coplanar helical configuration. This structure enables denser and more ordered molecular packing via steric repulsion, resulting in better spatial overlap between the expanded π-orbitals and the perovskite wavefunction¹⁵. As demonstrated in Fig. 2h, this characteristic facilitates more efficient hole extraction, reducing charge accumulation and thereby suppressing light-induced phase segregation. In contrast, 2PACz, with its carbazole functional group, exhibited inferior photo-stability due to its relatively less effective molecular packing and hole extraction properties. As a conventional hole transport material, TPD offers reasonable energy level alignment and hydrophobicity³⁹ but often forms non-uniform films on rough surfaces like CIGS⁴⁰. In contrast, SAMs spontaneously form conformal, ultrathin layers, ensuring better surface coverage on textured substrates and thereby improving interfacial contact and reducing trap states. As a result, SAM-based devices exhibit enhanced performance. Supplementary Figs. 13, 32 present the molecular structures^15,32,33, Quasi-Fermi Level Splitting (QFLS) analysis and pseudo J–V of the three HTLs obtained from the intensity-dependent V_OC, confirming that 4PADCB significantly mitigates V_OC loss and boosts device efficiency, supporting the use of NiO_x/4PADCB as an effective hybrid hole transport layer.
We further assessed the thermal stability of unencapsulated tandem devices stored in a nitrogen environment at different temperatures. After 6600 h of storage at 25 °C (Fig. 4b), the 4PADCB device retained 88.6% of its initial efficiency. In contrast, the Sput/4PADCB device demonstrated significantly improved stability, maintaining 98.7% of its initial efficiency. At 65 °C, after 678 h (Fig. 4c), the Sput/4PADCB device retained about 86% of its initial efficiency, slightly below the Sput/TPD device but significantly outperforming both 4PADCB and Sput/2PACz devices.
Maximum power point tracking (MPPT) measurements further reinforced these findings. The Sput/4PADCB device exhibited the highest stability, while the 4PADCB device showed the greatest degradation (Fig. 4d). The enhanced stability of NiO_x-containing devices is likely attributed to the abundant surface hydroxyl (OH) groups on NiO_x, which facilitate stronger molecular anchoring of SAM layers⁴¹. The superior stability of 4PADCB SAM over 2PACz SAM results from its ability to form denser and more uniform molecular packing.
In conclusion, the optimized AZO/Au/NiO_x/4PADCB IRL delivered exceptional PCE for perovskite/CIGS tandems. The combination of sputtered NiO_x and 4PADCB effectively eliminated the microscale shunt paths and efficiently passivated the interfacial traps. The ultrathin Au particle layer substantially boosted the recombination rate of the majority carriers from subcells. The efficient charge transport, coupled with strong molecular anchoring on NiO_x, enabled the perovskite/CIGS tandem devices to maintain robust long-term performance under elevated temperatures and prolonged light exposure. The principles behind this composite IRL design can be extended to other tandem architectures, offering a pathway toward the development of high-efficiency, stable, and durable multi-junction solar cells.
Formamidinium iodide (FAI), Cesium iodide (CsI), Lead iodide (PbI₂), Lead bromide (PbBr₂), Lead chloride (PbCl₂), Piperazinium Diiodide (PDI), Piperazine hydrobromide (PBr) and C₆₀ were obtained from Xi’an Polymer Light Technology in China. Dimethyl sulfoxide (DMSO), N, N-dimethylformamide (DMF), ethanol, and isopropanol (IPA) were purchased from Sigma-Aldrich. Diethyl ether (DE) was purchased from Jiangsu Yonghua in China. Poly(N, N’-bis(4-butylphenyl)-N, N-bis(phenyl)benzidine) (poly-TPD) was purchased from 1-Materials in Canada. 2PACz ([2-(9H-carbazol-9-yl)ethyl]phosphonic acid) was purchased from TCI. 4PADCB was synthesized by our team. Tetrakis(dimethylamino)tin(IV) (99.9999%) for atomic-layer-deposited (ALD) SnO₂ was bought from Suzhou Xinjiayuan Chemical Technology Co., Ltd. Indium tin oxide (ITO) and NiO_x targets were purchased from Shenzhen Zhongchengda Applied Materials Co., Ltd. Au(99.999%) was purchased from Hebei Fengming New Material Technology Co., Ltd. NiO_x nanoparticle powder (particle size around 10 nm) was purchased from Advanced Election Technology Company in China. ITO/glass substrates were purchased from Yiyang Huanan Xiangcheng Technology Co., Ltd, with a sheet resistance of 15 Ω/sq. Cu, In, Ga, and Se pellets were supplied by Kurt J. Lesker Company. Mo, intrinsic ZnO, and Al-doped ZnO targets were supplied by Huizhou Tianyi Rare Materials Co., Ltd.
The perovskite solar cell was fabricated in a p-i-n structure with ITO/4PADCB/perovskite/C60/BCP/Ag. First, a 4PADCB solution (0.5 mg/ml in ethanol) was spin-coated directly onto an ITO substrate at 3000 rpm for 30 s, followed by annealing at 100 °C for 10 min. Then, the 4PADCB-coated ITO sample was pre-washed by spin-coating a PBr solution (1.0 mg/ml) at a speed of 5000 rpm (with a ramping rate of 3000 rpm/s) for 30 s. Wide-bandgap perovskite solutions were prepared by dissolving FAI, CsI, PbI₂, PbBr₂, PbCl₂, with molar ratios adjusted to form (FA_0.8Cs_0.2)Pb(I_0.82Br_0.15Cl_0.03)₃ in a DMF and DMSO mixed solvent system (DMF: DMSO = 3:1 volume ratio). This solution was spin-coated onto the PBr film at 3000 rpm for 60 se. During this process, 500 μl of DE was quickly dripped onto the center of the perovskite film 30 s before the end of the spin-coating process, followed by annealing at 100 °C for 20 min. The surface passivation of the perovskite film was achieved by spin-coating a PDI solution (0.5 mg/ml). For the opaque devices, C₆₀ (20 nm), BCP (6 nm), and Ag (100 nm) layers were sequentially deposited onto the perovskite films by thermal evaporation.
The detailed CIGS and CdS preparation procedures are provided in Supplementary Figs. 1–4. Briefly, CIGS fabrication began with sputtering a ~1 μm molybdenum layer onto soda-lime glass as the back electrode. Then, Cu, In, Ga, and Se were co-evaporated via the multi-stage process to form a CIGS light absorber layer (~2.7 μm). The RbF post-deposition treatment was performed to improve the junction quality at the CIGS surface. The CIGS devices were completed with a ~50 nm CdS buffer layer via chemical bath deposition, a ∼50 nm i-ZnO layer, and a ~ 200 nm (for tandem devices) Al-doped ZnO window layer via RF sputtering. In the stand-alone configuration, Al grids and MgF₂ antireflection layers were additionally deposited by thermal evaporation.
For the fabrication of two-terminal tandem devices, the CIGS cells were terminated at the AZO layer. In the optimized scheme, a thin layer of 0.6 nm Au was deposited by thermal evaporation under high vacuum conditions (5 × 10⁻⁴ Pa). In other schemes, this step was skipped. Immediately afterward, 20 nm of NiO_x was sputtered in an O₂ (1%)/Ar (99%) mixed atmosphere at a process pressure of 0.5 Pa and a sputtering power of 60 W. An organic hole transport layer of either TPD, 2PACz, or 4PADCB was further spin-coated on the NiO_x layer. The light absorber layer of wide-bandgap perovskite was deposited in the same way as described above for the single-junction perovskite solar cells. It is worth noting that the concentration of the perovskite solution was now 1.80 mmol/mL, rather than the 1.35 mmol/mL concentration used for single-junction perovskite solar cells. To decrease parasitic absorption, a thinner C₆₀ layer (12 nm) was deposited by thermal evaporation, and a 20 nm thick SnO_x layer, instead of BCP, was deposited by reactive atomic layer deposition (ALD) (Picosun R200). The pulse and purge times were 0.3 s and 12 s, respectively, for TDMASn and 4 s and 15 s, for H₂O. The front transparent electrode of 100 nm thick ITO (In/Sn ratio 90/10) was sputtered in a pure argon atmosphere, and the frame electrode of 300 nm thick Ag was deposited by thermal evaporation. Finally, the MgF₂ antireflection layer (~100 nm) was deposited using electron beam evaporation.
J–V curves were measured using an AM 1.5 G solar simulator (Enli-tech AAA solar simulator, China) under ambient conditions with a scan speed of 0.1 V s⁻¹ (voltage steps of 20 mV and a delay time of 30 ms). The voltage range was from −0.1 V to 1.25 V for single-junction devices and −0.1 V to 2.0 V for tandem devices. The intensity of the solar simulator was calibrated to 1 sun (100 mW cm⁻²) using a reference Si solar cell. The boxes in the PV parameter plots indicate the 25/75 percentiles, and the whiskers mark the 10/90 percentiles. The line in the plots marks the respective average value.
The EQE curves were measured using a QE/incident photon-to-electron conversion efficiency (IPCE) system (Enli Technology Co. Ltd., China). The system was calibrated using a Si reference cell for the range from 300 to 1100 nm and a Ge reference cell for the range from 1100 to 1400 nm. For the tandem devices, LED bars were used as bias light sources, illuminating at a wavelength of 850 nm to measure the top cell and at a wavelength of 550 nm to measure the bottom cell.
The morphology and microstructure of the various samples were characterized using field-emission scanning electron microscopy (SEM, TESCAN MIRA3). Except for the Au particle on the conductive AZO sample, the remaining samples were coated with a thin gold layer (~5 nm, deposited for 30 s at 30 mA) to mitigate charging effects. SEM imaging was conducted under high vacuum, utilizing a beam energy of 15 kV.
GI-XRD analysis was carried out using a Rigaku SmartLab 3 kW diffractometer (Rigaku Co., Japan) with Cu Kα radiation, operated at a scanning speed of 5° min⁻¹.
The Raman spectrums were collected by a CCD detector (PIXIS256) through a spectrometer (Princeton Instrument, SpectraPro HRS-300) with a 1200 grooves/mm grating for a resolution of less than 1 cm⁻¹. The slit was fixed to 100 μm for 1 mi integration. A 532 nm laser was used to excite the samples through a 100(times) objective lens.
UPS measurements were performed on a Thermo Fisher ESCALAB Xi⁺ instrument, and the samples were placed in an ultrahigh vacuum. The measurements were conducted with an ultraviolet light source of He I (21.2 eV) with a pass energy of 0.5 eV, dwell time of 50 ms, bias voltage of −6 V, and a step size of 20 meV. The systematic energy error of UPS was ~50 meV. The relative accuracy of the work function and valence band maximum position was slightly better than 50 meV. UPS is very sensitive to the cleanliness of the film surface. To avoid contamination errors, we kept the freshly prepared samples in a nitrogen-filled box until they entered the UPS chambers. XPS measurements were performed on a Thermo Fisher ESCALAB Xi⁺ instrument with a monochromatic Al Kα (1486.6 eV) X-ray, and the samples were measured under an ultrahigh vacuum. The binding energies were calibrated using the C(1 s) carbon peak (284.8 eV). All the high-resolution spectra were obtained under constant analyzer energy mode with a pass energy of 30 eV and a step size of 0.1 eV.
AFM height images and KPFM surface potential distribution images were obtained under ambient conditions using a Bruker Dimension Icon AFM. These measurements employed amplitude modulation mode, using SCM-PIT-V2 probes from Bruker, which have a resonance frequency of 70 kHz and a force constant of 3 N/m. A consistent lift height of 20 nm was maintained throughout the measurements. Regarding the measurement under nitrogen atmosphere, we maintain the same testing conditions and use a custom-built acrylic box with continuous nitrogen purging to prevent ambient degradation for KPFM measurements.
KPFM measurements directly provided the spatially resolved contact potential difference (CPD), which was defined as
where Φ_tip and Φ_sample were the work functions of the tip and sample, respectively, and e was the elementary charge. When a standard reference, such as highly oriented pyrolytic graphite (HOPG) with a known work function, was used, the absolute work function of samples could be determined by
The PL spectra of the perovskite films were recorded using a DeltaFlex fluorescence spectrometer (HORIBA, Japan), excited with a semiconductor laser at a center wavelength of 488 nm. TRPL was carried out with a time-correlated single-photon counting (TCSPC) module as the detector. Carrier lifetime is measured using TRPL decay curves. The fitting parameters, obtained from a bi-exponential decay equation, are listed below:
where ({tau }_{{{rm{ave}}}}) is the average carrier lifetime, ({A}_{1}) and ({A}_{2}) represent the decay amplitudes of fast and slow decay processes, respectively, and ({tau }_{1}) and ({tau }_{2}) stand for the fast and slow decay time constants, respectively.
The FLIM measurement was performed with the confocal configuration of a PicoQuant MicroTime 200 instrument. The excitation pulsed laser at 488 nm was operated at a repetition rate of 5 MHz and was focused onto the sample through an objective lens (Olympus MPLFLN100x, NA 0.9). The fluorescence emission of the sample was passed through a 650 nm long-pass filter (Jcoptix, OFE1LP-650) and detected by a single-photon avalanche diode (SPAD). The electrical signal was then processed by time-correlated single-photon counting (TCSPC) electronics (Time Harp 260, PicoQuant). The FLIM images were analyzed by PicoQuant SynphoTime 64.
The Hall measurements were characterized by the Hall effect tester (HMS-5500, Ecopia). Place the sample in the testing module at room temperature and atmospheric conditions to obtain information on the sheet resistance, resistivity, and carrier concentration of the thin film.
Photostability measurements were performed using a home-built setup. The perovskite films were excited by a 405 nm continuous-wave laser, and the illumination intensity was adjusted to 1-, 10-, and 30-sun-equivalent intensity using a variable neutral density filter (LBTEK, NDFR-50S-3M). Each measurement lasted a total of 20 min, with the fluorescence spectrum collected every minute to characterize the sample’s stability under illumination. The PLQY measurements were characterized by the luminescence test system (Enli Technology Co. Ltd, China, LQ-100), with an excitation wavelength of 365 nm.
Measurements were conducted at 10 Hz in an ambient environment using a 405 nm nanosecond pulsed diode laser (Thorlabs, NPL41C) with a pulse duration of 6 ns and an intensity of approximately 0.75 μJ/cm², triggered by an arbitrary function generator (Tektronix, AFG31000). The tr-SPV signal was measured using an oscilloscope (Tektronix, MDO34, 2.5 GS/s). The measurement frequency was 10 Hz, and 128 averages were taken per routine. Meanwhile, to avoid atmospheric impact, we conducted SPV measurements under nitrogen atmosphere using a custom-built acrylic box with continuous nitrogen purging to prevent ambient degradation, while ensuring that other testing conditions were the same. The sample was excited from the ITO side, and the SPV signals were collected using two copper wires, which connected the Au electrode on the front side and the ITO electrode on the back side of the half device. These open-circuit signals were then transferred to the oscilloscope through a BNC adapter. Carrier extraction time is measured using tr-SPV curves. It is defined as the time required to reach the maximum SPV. The decay time is determined by fitting the data to a bi-exponential decay equation, as described below:
where ({V}_{10}) and ({V}_{20}) represent the decay amplitudes of fast and slow decay processes, respectively, and ({tau }_{1}) and ({tau }_{2}) stand for the fast and slow decay time constants, respectively.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
The main data supporting the findings of this study are available within the published article and its Supplementary Information and source data files. Additional data are available from the corresponding author on request. Source data are provided with this paper.
Liu, J. et al. Perovskite/silicon tandem solar cells with bilayer interface passivation. Nature 635, 596–603 (2024).
Article  CAS  PubMed  ADS  Google Scholar 
Aydin, E. et al. Enhanced optoelectronic coupling for perovskite/silicon tandem solar cells. Nature 623, 732–738 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
Chin, X. Y. et al. Interface passivation for 31.25%-efficient perovskite/silicon tandem solar cells. Science 381, 59–62 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
Hou, Y. et al. Efficient tandem solar cells with solution-processed perovskite on textured crystalline silicon. Science 367, 1135–1140 (2020).
Article  CAS  PubMed  ADS  Google Scholar 
Aydin, E. et al. Interplay between temperature and bandgap energies on the outdoor performance of perovskite/silicon tandem solar cells. Nat. Energy 5, 851–859 (2020).
Article  CAS  ADS  Google Scholar 
Han, Q. F. et al. High-performance perovskite/Cu(In,Ga)Se₂ monolithic tandem solar cells. Science 361, 904 (2018).
Article  CAS  PubMed  ADS  Google Scholar 
Liu, X. X. et al. Over 28% efficiency perovskite/Cu(InGa)Se₂ tandem solar cells: highly efficient sub-cells and their bandgap matching. Energ. Environ. Sci. 16, 5029–5042 (2023).
Article  CAS  ADS  Google Scholar 
Kafedjiska, I. et al. Advanced characterization and optimization of NiO_x:Cu-SAM hole-transporting Bi-Layer for 23.4% efficient monolithic Cu(In,Ga)Se₂-perovskite tandem solar cells. Adv. Funct. Mater. 33, 2302924 (2023).
Article  CAS  Google Scholar 
Jost, M. et al. Albrecht, 21.6%-efficient monolithic perovskite/Cu(In,Ga)Se₂ tandem solar cells with thin conformal hole transport layers for integration on rough bottom cell surfaces. ACS Energy Lett. 4, 583–590 (2019).
Article  CAS  Google Scholar 
Al-Ashouri, A. et al. Conformal monolayer contacts with lossless interfaces for perovskite single junction and monolithic tandem solar cells. Energ. Environ. Sci. 12, 3356–3369 (2019).
Article  CAS  Google Scholar 
Jost, M. et al. Perovskite/CIGS tandem solar cells: from certified 24.2% toward 30% and beyond. ACS Energy Lett 7, 1298–1307 (2022).
Article  CAS  Google Scholar 
Lin, R. X. et al. All-perovskite tandem solar cells with 3D/3D bilayer perovskite heterojunction. Nature 620, 994–1000 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
Zhou, S. et al. Aspartate all-in-one doping strategy enables efficient all-perovskite tandems. Nature 624, 69–73 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
Xiao, K. et al. Scalable processing for realizing 21.7%-efficient all-perovskite tandem solar modules. Science 376, 762–767 (2022).
Article  CAS  PubMed  ADS  Google Scholar 
He, R. et al. Improving interface quality for 1-cm² all-perovskite tandem solar cells. Nature 618, 80–86 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
National Renewable Energy Laboratory. Best Research-Cell Efficiency Chart. https://www.nrel.gov/pv/cell-efficiency.html (2025).
Green, M. A. et al. Solar cell efficiency tables (Version 64). Prog. Photovolt: Res. Appl. 32, 425–441 (2024).
Article  Google Scholar 
Leijtens, T., Bush, K. A., Prasanna, R. & McGehee, M. D. Opportunities and challenges for tandem solar cells using metal halide perovskite semiconductors. Nat. Energy 3, 828–838 (2018).
Article  CAS  ADS  Google Scholar 
Yoon, G. W., Jo, B., Boonmongkolras, P., Han, G. S. & Jung, H. S. Perovskite tandem solar cells for low Earth orbit satellite power applications. Adv. Energy Mater. 15, 2400204 (2024).
Article  Google Scholar 
Yamaguchi, M. Radiation resistance of compound semiconductor solar cells. J. Appl. Phys. 78, 1476–1480 (1995).
Article  CAS  ADS  Google Scholar 
Ruiz-Preciado, M. A. et al. Monolithic two-terminal perovskite/CIS tandem solar cells with efficiency approaching 25%. ACS Energy Lett 7, 2273–2281 (2022).
Article  CAS  PubMed  PubMed Central  Google Scholar 
Jeong, I. et al. Flexible and lightweight perovskite/Cu(In,Ga)Se₂ tandem solar cells. Joule 9, 101794 (2025).
Article  CAS  Google Scholar 
Tian, L. et al. Divalent cation replacement strategy stabilizes wide-bandgap perovskite for Cu(In,Ga)Se₂ tandem solar cells. Nat. Photonics 19, 479–485 (2025).
Article  CAS  ADS  Google Scholar 
Tang, L. et al. Record-efficient flexible monolithic perovskite-CIGS tandem solar cell with V_OC exceeding 1.8 V on polymer substrate. Adv. Energy Mater. 15, 2403682 (2025).
Article  CAS  ADS  Google Scholar 
Feng, Z. et al. Doped-NiO_x seed layer on textured substrates for low-loss contacts in perovskite solar cells. Adv. Energy Mater. 15, 2405016 (2024).
Article  Google Scholar 
Liu, Y. et al. Co-adsorbent boosting the performance of perovskite solar cell based on hole-selective self-assembled molecules. Adv. Funct. Mater. 35, 2421576 (2025).
Article  CAS  Google Scholar 
Jiang, W. et al. π-expanded carbazoles as hole-selective self-assembled monolayers for high-performance perovskite solar cells. Angew. Chem. Int. Ed. Engl 61, 1–6 (2022).
Article  Google Scholar 
Islam, M. B., Yanagida, M., Shirai, Y., Nabetani, Y. & Miyano, K. NiO_x hole transport layer for perovskite solar cells with improved stability and reproducibility. ACS Omega 2, 2291–2299 (2017).
Article  CAS  PubMed  PubMed Central  Google Scholar 
Azmi, R. et al. Double-sided 2D/3D heterojunctions for inverted perovskite solar cells. Nature 628, 93–98 (2024).
Article  CAS  PubMed  ADS  Google Scholar 
Zhang, S. et al. Minimizing buried interfacial defects for efficient inverted perovskite solar cells. Science 380, 404–409 (2023).
Article  CAS  PubMed  ADS  Google Scholar 
Zhang, J., Yu, B., Sun, Y. & Yu, H. Minimized energy loss at the buried interface of p-i-n perovskite solar cells via accelerating charge transfer and forming p-n homojunction. Adv. Energy Mater. 13, 2300382 (2023).
Article  CAS  Google Scholar 
Al-Ashouri, A. et al. Monolithic perovskite/silicon tandem solar cell with >29% efficiency by enhanced hole extraction. Science 370, 1300–1309 (2020).
Article  CAS  PubMed  ADS  Google Scholar 
Wang, H. et al. A multi-functional molecular modifier enabling efficient large-area perovskite light-emitting diodes. Joule 4, 1977–1987 (2020).
Article  CAS  Google Scholar 
De Bastiani, M. et al. Recombination junctions for efficient monolithic perovskite-based tandem solar cells: physical principles, properties, processing and prospects. Materials Horizons 7, 2791–2809 (2020).
Article  Google Scholar 
Zhang, J. et al. Highly efficient narrow bandgap Cu(In,Ga)Se₂ solar cells with enhanced open circuit voltage for tandem application. Nat. Commun. 15, 10365 (2024).
Article  CAS  PubMed  PubMed Central  ADS  Google Scholar 
Hosseinian Ahangharnejhad, R. et al. Heben, Irradiance and temperature considerations in the design and deployment of high annual energy yield perovskite/CIGS tandems. Sustainable Energy & Fuels 3, 1841–1851 (2019).
Article  CAS  Google Scholar 
Mahesh, S. et al. Revealing the origin of voltage loss in mixed-halide perovskite solar cells. Energ. Environ Sci. 13, 258–267 (2020).
Article  MathSciNet  CAS  Google Scholar 
Slotcavage, D. J., Karunadasa, H. I. & McGehee, M. D. Light-induced phase segregation in halide-perovskite absorbers. ACS Energy Lett 1, 1199–1205 (2016).
Article  CAS  Google Scholar 
Zhao, X. & Tan, Z.-K. Large-area near-infrared perovskite light-emitting diodes. Nat. Photonics 14, 215–218 (2019).
Article  ADS  Google Scholar 
Magomedov, A. et al. Self-assembled hole-transporting monolayer for highly efficient perovskite solar cells. Adv. Energy Mater. 8, 1801892 (2018).
Article  Google Scholar 
Kuan, C. H. et al. Functionalized thienopyrazines on NiO_x film as a self-assembled monolayer for efficient tin-perovskite solar cells using a two-step method. Angew. Chem. Int. Ed. 136, 1–12 (2024).
Article  Google Scholar 
Download references
This work was supported by the National Key Research and Development Program of China (Grant No. 2024YFB4205300 to J.L. and X.X.), the Special Fund for the “Dual Carbon” Science and Technology Innovation of Jiangsu province (Industrial Prospect and Key Technology Research Program, BE2022021 to X.X.), the National Natural Science Foundation of China (Grant No. 61904128 to J.G. and No. 12374357 to S.W.), and the Knowledge Innovation Program of Wuhan-Shuguang Project (Grant No. 2023010201020269 to J.L. and 2023020201020253 to J.G.). The authors gratefully acknowledge the support provided by Wuhan University’s start-up fund, as well as the facilities and assistance offered by the Core Facility of Wuhan University. Special thanks are extended to Prof. Guojia Fang, Mr. Hongyi Fang, and Mr. Guoyi Chen (Wuhan University) for their assistance with PL and PLQY measurements, and to Prof. Chao Chen and Mr. Zhihan Ding (Huazhong University of Science and Technology) for their support with Hall measurements.
These authors contributed equally: Wang Li, Junjun Zhang, and Li Zeng.
School of Physics and Technology, Key Lab of Artificial Micro- and Nano-structures of Ministry of Education, Wuhan University, Wuhan, 430072, Hubei, China
Wang Li, Junjun Zhang, Li Zeng, Zhou Fang, Xinxing Liu, Zengyang Ma, Yitian Zhang, Hui Yan, Chen Shen, Zhuo Xue, Jingyi Zhu, Qiren Luo, Chang Liu, Jianmin Li, Sheng Wang, Junbo Gong & Xudong Xiao
Institute of Flexible Electronics (IFE Future Technologies), College of Materials, Innovation Laboratory for Sciences and Technologies of Energy Materials of Fujian Province (IKKEM), Xiamen University, Xiamen, 361005, China
Wanhai Wang & Weihua Tang
State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, Wuhan, 430070, PR China
Ruixuan Jiang & Tongle Bu
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
X.X. supervised the project, provided the primary funding, and offered critical guidance on experimental design and manuscript preparation. X.X., J.G., S.W., and J.L. conceived the overall idea, discussed the entire experimental process, and wrote and polished the manuscript. All authors contributed to the review of the manuscript. During the revision process, J.L., W.L., and X.X. made significant contributions to refining the manuscript structure, strengthening data interpretation, enhancing the clarity and scientific rigor of the presentation, and formulating detailed and thoughtful responses to the reviewers’ comments. W.L., J.J.Z., and L.Z. contributed equally to this work. They designed and conducted the primary experiments, fabricated the tandem devices, and performed most of the characterization. J.G. consulted and coordinated the acquisition of 4PADCB. W.W. and W.T. synthesized and provided the 4PADCB material. Z.F. assisted in the fabrication of perovskite solar cells. X.L. contributed to the optimization of wide-bandgap perovskite cells. Z.M. and Y.Z. supported the maintenance of CIGS fabrication equipment and optimized the AZO layer thickness. H.Y. and J.L. improved the surface roughness of the CIGS bottom cells. C.S., Z.X., and S.W. performed FLIM, AFM, TRPL, KPFM, and PL stability measurements. J.Y.Z. and S.W. carried out Tr-SPV measurements. Q.L. and C.L. provided resources for ALD and assisted in the preparation of SnO₂ functional layers. R.J. and T.B. conducted PL measurements and contributed to the device stability test.
Correspondence to Jianmin Li, Sheng Wang, Junbo Gong or Xudong Xiao.
The authors declare no competing interests.
Nature Communications thanks the 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 4.0 International License, which permits use, sharing, adaptation, 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 changes were made. 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/4.0/.
Reprints and permissions
Li, W., Zhang, J., Zeng, L. et al. Efficient perovskite/Cu(In,Ga)Se₂ tandem solar cells with a composite intermediate recombination layer. Nat Commun 17, 711 (2026). https://doi.org/10.1038/s41467-025-67350-y
Download citation
Received: 13 October 2025
Accepted: 27 November 2025
Published: 10 December 2025
Version of record: 20 January 2026
DOI: https://doi.org/10.1038/s41467-025-67350-y
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
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.

source