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: 2093 (2026)
9556
2
15
Metrics details
Wide-bandgap perovskite solar cells are essential for constructing multi-junction solar cells; nevertheless, their achievable photovoltage is often limited by non-radiative recombination losses caused by defect states, mismatched energy levels, and poor contact at interfaces, resulting in a photovoltage plateau beyond 1.68 eV. Here, we explore intricate Lewis acid-base interactions between hole-selective self-assembled monolayer and ammonium ligands to engineer a localized 2D/3D perovskite heterojunction at the buried interface. This structure not only help reduce defect density but also facilitate charge extraction and templated perovskite crystallization without compromising the bulk optoelectrical properties of 3D perovskite. These resulted in 1.30, 1.38, and 1.42 V photovoltages for 1.68, 1.79, and 1.85 eV bandgap perovskite solar cells, respectively, all exceeding 90% of their thermodynamic limits. We also demonstrate high-efficiency monolithic tandem solar cells by integrating the optimized wide-bandgap perovskite solar cell with an organic subcell to achieve a PCE of 27.11% (certified 26.3%).
Tandem solar cells (TSCs) offer a viable route to surpass theoretical efficiency limits in single-junction solar cells1,2,3. Among these, wide-bandgap (WBG) perovskites are promising for use as front sub-cells to match various narrow-bandgap (NBG) sub-cells, owing to their high tunability of bandgap by mixing halides (i.e., I, Br, and Cl)4,5. However, the substantial non-radiative recombination loss in WBG perovskites remains a critical challenge for gaining higher efficiency in perovskite-based TSCs6,7. It mainly arises from the high density of interfacial defects, and low charge extraction efficiency resulted from sub-optimal energy level between charge transport/selective layer (CTL/CSL) and perovskites, exacerbated by high bromide content (> 20%)8,9,10,11. These collective challenges contribute to a large photovoltage deficit and poor long-term operational stability.
Over the past decade, various strategies have been reported to reduce surface defect density for minimizing photovoltage losses3,12,13,14. One particular approach involves the use of low-dimensional perovskites (LDPs) as localized passivation layer on top of perovskite surface, which has driven remarkable advances in power conversion efficiency (PCE) for perovskite solar cells (PSCs). Given that most LDPs exhibit a Fermi level close to their valence band maximum (VBM), they are ideal interfacial layers for a hole-selective SAM/perovskite buried interface to facilitate hole extraction15. Nevertheless, it is challenging to spatially confine LDPs at the buried interface, despite it will have a greater impact on device performance. The reported approaches generally lead to random LDP distribution and diffusion (i.e., from buried interface into 3D perovskite bulk) because of insufficient molecular interactions at the buried interface, exacerbating intrinsic quantum confinement effects and anisotropic charge transport16,17,18.
Hole-selective self-assembled monolayers (SAMs) have been demonstrated to play a critical role in achieving high PCE in p-i-n structured PSCs by facilitating fast hole extraction and providing effective passivation on metal oxide surfaces (ITO or FTO). Previously, we introduced a helical SAM with extended π-conjugation, named CbzNaph, which is advantageous for both PSCs and TSCs owing to its enhanced dipole moment and intermolecular π-π interactions19,20. However, the extended conjugation elevates the highest occupied molecular orbital (HOMO) energy level of CbzNaph, thus enlarges the energetic mismatch with the valence band maximum (VBM) of WBG perovskites. Besides, its all-carbon-hydrogen structure at the apex of the molecular skeleton lacks the needed functionalities for inducing molecular interactions with the upper layer.
In this work, we demonstrate the successful realization of a localized 2D/3D heterojunction at the SAM/perovskite buried interface by taking advantage of Lewis acid-base interactions between hole-selective SAM and ammonium ligands. We design and synthesize a SAM molecule, named CbzBT-B, with sulfur (S) heteroatom on its head group to facilitate its critical interaction with the -NH3+ group in the oleylammonium iodide (OAmI) ligand, to confine 2D/3D perovskite heterojunction. The localized heterojunction can passivate the buried interface without compromising the optoelectrical properties of the perovskite bulk phase. It also templates the perovskite growth, leading to high-quality films with preferred orientation21. Meanwhile, the rigid phenyl linker in the CbzBT-B enhances its stability and the electron delocalization within the skeleton to lower its HOMO level for better aligning with WBG perovskites20,22. The resulting 1.79 eV-PSC achieves a PCE of 20.23% with an open-circuit voltage (VOC) of 1.382 V, corresponding to approximately 92% of its thermodynamic limit. Integration of this optimized WBG PSC with an organic rear subcell yields a high PCE of 27.11% (certified 26.3%) for a monolithic perovskite-organic TSC (PO-TSC).
Various strategies have been reported to construct 2D/3D heterojunctions at the buried interface, primarily involving the pre-deposition of ammonium ligands on CTLs/CSLs23,24,25. However, due to weak interfacial interactions, a significant portion of ligands tends to diffuse into the perovskite bulk, resulting in randomly distributed 2D perovskites throughout the 3D perovskite bulk (Fig. 1 and Supplementary Fig. 1). Some ligands even accumulated at the buried interface because of self-aggregation, generating many voids at the interface (Supplementary Fig. 2a). Moreover, applying solvent (dimethylformamide, DMF) washing (w/ SW) on the as-formed ligand layer results in only trace ligand residues with an uncontrolled (i.e., sporadic) distribution on the substrate (Supplementary Fig. 2b).
The construction of buried interfaces through various strategies (The orange stripes represent 2D perovskite, the reddish polygons indicate aggregates or residues of ligands and the stripes near the ITO are corresponding SAM, respectively).
To tackle these problems, we explored the intricate Lewis acid-base interaction between ammonium ligands and a rationally designed SAM to localize them for forming a 2D/3D heterojunction at the buried interface. The molecular structures of CbzBT-based SAMs are illustrated in Fig. 2a, with detailed synthesis and material characterizations provided in the Supplementary Information. The Lewis basic S heteroatoms within the head groups are the key to enabling intimate interactions with Lewis acidic ammonium ligands, thereby localizing ligands to facilitate the formation of LDPs at the buried interface26,27. Specifically, the ligand layer was pre-deposited on CbzBT, followed by solvent washing to remove excess molecules. Moreover, the asymmetric benzothiophene unit incorporated in the carbazole framework downshifted its HOMO level to −5.54 and −5.72 eV for CbzBT with alky (denoted as CbzBT-C) and benzene (denoted as CbzBT-B) linkers, respectively (Fig. 2a and Supplementary Figs. S3–4). This modification also enhanced the molecular dipole moment to exceed 2.6 D. The phenyl linker in CbzBT-B facilitates electron delocalization within the skeleton, potentially mitigating molecular instability arising from the coexistence of electron-rich carbazole and benzothiophene units20,22.
a Molecular structure and calculated HOMO energy levels of CbzNaph, CbzBT (C4) and CbzBT (benzene), respectively. b High-voltage cyclic voltammetry (CV) test of corresponding SAM, Insets: photographs of SAM materials solution after ageing. c XPS spectra of S 2p for the corresponding films. d Statistical distribution of surface potential extracted from the KPFM images. TOF-SIMS profile of perovskites on ITO/ ligand modified CbzBT-B (e) without and (f) with solvent-treatment. g XRD patterns of the exposed buried interface of the corresponding films (Shadings is indicate of LDPs region).
We first performed high-voltage cyclic voltammetry (CV) and continuous ultraviolet (UV) irradiation at 90 °C to assess the electrical and photo-stability of the fused CbzBT compounds, as shown in Fig. 2b, Supplementary Fig. 5 and Supplementary Note 1. The results showed prominent signal attenuation for CbzBT-C while CbzBT-B exhibited minimal signal attenuation, verifying the enhanced molecular stability with the benzene linker. Consequently, CbzBT-B was selected for studying its Lewis acid-base interaction with ammonium ligands.
To verify the interactions between S atoms on the CbzBT-B skeleton and the ligands, we compared the chemical shift of the ammonium ligand when mixed with CbzBT-B and CbzNaph (as control sample) using proton nuclear magnetic resonance (1H NMR) spectroscopy (Supplementary Fig. 6). The α-H of the ligand exhibited a pronounced chemical shift when blended with CbzBT, demonstrating the presence of Lewis acid-base interactions. High-resolution X-ray photoelectron spectroscopy (HR-XPS) was also conducted to further study these interactions in the form of ITO/CbzNaph or CbzBT-B/ligand thin films. As shown in Supplementary Fig. 7, negligible I 3 d electron signals (binding energies around 630.5 and 619.0 eV) were observed for the OAmI deposited on CbzNaph after solvent washing. In contrast, the peak from the I 3 d was reduced somewhat but remained relatively prominent for the OAmI prepared on CbzBT-B, implying successful ligand retention. Meanwhile, XPS analysis shows the S 2p1/2 and 2p3/2 binding energies shifted from 165.06 and 163.89 eV to 165.17 and 163.97 eV, respectively (Fig. 2c), confirming the interaction between CbzBT-B and the ligand.
Then, the surface potential profile was investigated using Kelvin probe force microscopy (KPFM) to assess the impact of ligand modification (Fig. 2d and Supplementary Figs. 8–9). Initially, both SAMs exhibited similar potential shifts upon ligand deposition. However, solvent washing revealed a distinct contrast: CbzNaph films showed a pronounced potential shift of Δφ = 312 mV, while CbzBT-B retained 63% of its initial shift (Δφ = 198 mV), demonstrating effective ligand anchoring on the CbzBT-B (Fig. 2d). Meanwhile, the results from atomic force microscopy (AFM) exhibited apparent ligand aggregation in both samples, with CbzNaph showing a notably reduced surface roughness after solvent washing (Supplementary Figs. 8–9). These observations corroborated the interaction between CbzBT-B and ligand, resulting in the ligand retained at the buried interface while excess ligands were removed after solvent washing.
We further employed time-of-flight secondary ion mass spectrometry (ToF-SIMS) to investigate the out-of-plane distribution of ligands. For the unwashed samples, a portion of the ligand was observed to diffuse from the buried interface into the perovskite bulk (Fig. 2e–f and Supplementary Fig. 1). The buried surface of perovskite also showed numerous voids as characterized by top-viewing scanning electron microscopy (SEM), possibly resulting from the self-aggregated ligands (Supplementary Fig. 2), consistent with that observed in the AFM result (Supplementary Figs. 8–9). The presence of the white flakes (outlined by dashed circles) at the buried interface is indicative of agglomerated PbI2. In contrast, the perovskite prepared on the solvent-washed ligand layer exhibited negligible ligand assembly and PbI2 aggregation. Thin-film X-ray diffraction (XRD) was used to confirm the 2D perovskite formation at the buried interfaces, as illustrated in Fig. 2g. The samples prepared on ligand-coated CbzBT-B exhibited pronounced diffraction peaks from LDPs at 2θ ~ 7.2°, which can be assigned to LDPs (Supplementary Fig. 10a). The weaker diffraction peak of LDPs from CbzNaph-based samples suggests the inhomogeneous LDP distribution, with these peaks disappearing after solvent washing. These results validate our strategy in effectively facilitating localized LDP formation. Moreover, the reduced PbI2 peak in CbzBT-B (w/SW)-based perovskite further supports its effectiveness in regulating buried interface characteristics, aligning with those observed by SEM (Supplementary Fig. 2).
To gain more insights on the LDP distribution, grazing-incidence wide-angle X-ray scattering (GIWAXS) with two incident angles (α) of 0.1° and 0.3° was conducted for the samples28,29. As presented in Fig. 3a–b and S11, no LDP signal was detected at either depth for the perovskite prepared on CbzNaph/ligand, indicating that most of the ligand was washed away by the solvent, and thus cannot form LDPs. For the CbzBT-B-based sample, the diffraction qz peak at ~ 0.3 Å-1 can be observed when the incident angle is increased to 0.3°, implying that the 2D perovskites were predominantly formed at the buried interface. This is further consolidated by the excitonic emission peak of LDPs photoluminescence (PL) spectrum from the bottom side of the perovskite prepared on CbzBT-B/ligand (w/SW), which should result from the LDPs’ quantum confinement effect at the buried interface (Fig. 3c and Supplementary Fig. 12)28,29,30. To explicitly identify the spatial distribution and morphology of LDPs, we have conducted SEM combined with energy dispersive X-ray spectroscopy (EDX) for the samples using FPEAI ligands (which contain F as distinct elemental markers), shown in Supplementary Fig. 13. The buried interface SEM and EDX show that LDPs are mainly distributed along grain boundaries. Based on the EDX results and the presence of the LDPs characteristic peak in XRD and GIWAXS data, we propose the following formation mechanism of LDPs during the crystallization process: as perovskite grains grow and coalesce, most ligands are expelled to evolving grain boundaries to form confined LDPs. A smaller fraction of ligands remains within the grain, passivating intrinsic vacancy defects. This scenario explains both the ~10 nm interface-localized layer and its discontinuous, island-like morphology. Notably, the CbzBT-B-based sample exhibits (100) dominated facet with preferential orientation perpendicular to the substrate (Fig. 3d), attributed to the templating effect of the localized 2D/3D heterojunction. Compared with other facets, the (100) facet usually exhibits improved carrier mobility and reduced trap densities than other facet orientations, as proven previously by both theoretical and experimental studies31.
a–b GIWAXS mappings of the CbzNaph-w/ SW and CbzBT-B-w/ SW perovskite films with different incidence angles. c Normalized PL spectra with different excitation directions from the glass side and top perovskite surface film of corresponding films. Inset: magnified PL spectra of samples. d The logarithmic intensity of perovskite crystal plane corresponds to a q ≈ 1.0 Å−1 plotted along the polar angle. e–f In-situ PL peak intensity and position evolution of corresponding perovskite films during spin-coating (dash-line indicates the peak position evolution of the film during spinning). g In-situ PL peak intensity evolution for corresponding perovskite films during the annealing process.
To elucidate the effect of localized ligands on the formation of the upper 3D perovskite, we performed in-situ PL spectroscopy to track nucleation and crystallization dynamics during film growth. The control and target perovskite films were prepared on CbzBT-B with and without pre-deposited ligand (w/SW), respectively. Upon antisolvent dripping, both samples exhibited rapid PL shifts to approximately 660 nm, representing the onset of nucleation (Fig. 3e–g). The target film maintained a stable PL profile, whereas the PL of the control sample gradually blue-shifted and its nucleation was dominated by Br-rich components after 24 s into the spin-coating process (Fig. 3f). This phenomenon can be attributed to the lower formation energy of Br-rich nuclei, as revealed by previous studies32. Throughout the crystal growth stage, the PL peak of the two samples evolved along comparable trajectories within 40 s (Fig. 3g and Supplementary Fig. 14)32,33. The solvent evaporation facilitated rapid I– ion incorporation into the Br-rich perovskite lattice, resulting in the target perovskite with the designed formula. As depicted in Fig. 3g, crystal growth presented a temporarily increased PL intensity, followed by a notable decrease due to the formation of grain boundaries and thermal effects (thermally activated carrier traps and lattice vibration caused by continuous heating). It is noted that the target perovskite film undergoes rapid crystal growth with a distinct Ostwald ripening, achieving higher PL intensity34. Therefore, these results indicate that the cations at the buried interface govern the relative nucleation kinetics of I- and Br-containing domains and facilitate secondary growth for enhanced crystallinity33. The homogeneity was visualized through PL mapping, with the target film showing a narrower emission distribution (FWHM = 40 nm at 692 nm) compared to the control (FWHM = 45 nm at 694 nm), confirming improved uniformity of crystallization induced by localized ligand (Supplementary Fig. 15).
To evaluate the thin film properties of the as-formed perovskites, their conductivities were measured, as illustrated in Fig. 4a and Supplementary Fig. 16. The samples based on unwashed ligands exhibited lower conductivities with fluctuations because of the uncontrollable diffusion of ligands into the 3D perovskite bulk. Notably, the conductivity of the CbzNaph-based films exhibited an obvious increase after solvent washing, suggesting that most buried ligands were removed from the substrate due to the absence of chemical interactions. We performed ultraviolet photoelectron spectroscopy (UPS) measurements to assess the impact of LDP distribution on the electronic structure and optoelectrical behavior of the perovskite films. As depicted in Fig. 4b and S17, the top surface of perovskite grown on ITO/CbzBT-B/ligand (w/o SW) exhibits less n-type characteristics compared with that of the washed sample, illustrating that the excessive LDPs in the bulk phase are undesirable for electron extraction at the top interface. Supplementary Fig. 17c depicts the energetic alignment of perovskite absorbers prepared on different substrates. The HOMO levels are −5.38 eV and −5.71 eV for CbzNaph and CbzBT-B, respectively, suggesting better energetic alignment between CbzBT-B and WBG perovskite absorbers. Meanwhile, the top surface of the target perovskite shows a more n-type characteristic, probably resulting from a considerable number of electron-filled traps. These favorable energy alignments are conducive to charge extraction, therefore mitigating interfacial trap-assisted recombination35.
a Film conductivity of corresponding wide-bandgap perovskite films (Error bar: standard deviation). b Schematic representation of energy level alignment of corresponding films based on values from UPS measurements. c–d Transient absorption measurements of control and target films showing the transient reflectivity change (measured in optical density) over a timescale of 6000 ps. e Intensity quenching of the main peak of TAS. f–g Dependence of the trap density on the profiling distance of corresponding films. h Statistics of mobilities and defect density for the corresponding films (Error bar: standard deviation).
Time-resolved photoluminescence (TRPL) results of the control and target perovskite (with localized 2D/3D heterojunction) films deposited on the CbzBT-B/ITO substrates were used to analyze the charge dynamics (Supplementary Fig. 18 and Supplementary Note 2). Compared with the control sample (τavg = 40 ns), the target perovskite exhibited a prolonged τavg of 104 ns, implying reduced non-radiative recombination within the film36. The transient absorption spectroscopy (TAS) provides more insights into the impact of the localized heterojunction on carrier dynamics, as illustrated in Fig. 4c–e. Both perovskites show distinct ground-state bleach (GSB) peaks at approximately 670 nm, where the carrier lifetimes (Fig. 4e) were obtained by fitting the TAS curves with a three-exponential decay function (Supplementary Note 3). The target film displays reduced τ1 and τ2 values (0.29 ps and 160 ps) in comparison to the control film (0.32 ps and 170 ps), indicating more expedient trap filling and potentially improved VOC in PSCs37,38. These results are consistent with the TRPL that demonstrate the enhanced hole extraction at the buried interface in the presence of localized 2D/3D heterojunction. It is noted that there is no GSB signal from the excitons of 2D perovskites, suggesting that the localized 2D phase merely serves as a conduit for carrier extraction and transport.
To probe how localized 2D/3D heterojunctions influence defect passivation, we applied a drive-level capacitance profiling method (DLCP) to measure vertical distribution (with respect to substrate) of trap density (Fig. 4f–g) at the device level. The trap densities at the buried interface are more than three orders of magnitude higher than that on the top surface for control perovskite. In contrast, the spatial distribution of trap density in the target sample exhibited no distinct difference at both sides, demonstrating the effective passivation on the buried interface. Then, trap-state densities and charge-carrier mobilities were extracted via the space-charge-limited current (SCLC) method, and the fitting results presented in Supplementary Fig. 19. Compared with the control perovskite, the hole trap density (Nt) of the target film was reduced from 2.75 × 1016 to 6.23 × 1015 cm-3 (Fig. 4h). The hole mobility of target perovskite increased from 1.11 × 10−4 to 1.42 × 10−4 cm2 V−1 s−1 with a smaller standard deviation. We ascribed the notably reduced density of trap states and enhanced charge mobility to the passivation and templating effect of localized heterojunction in the target perovskite, respectively33.
Based on these encouraging results, single-junction perovskite solar cells were prepared using the stack glass/ITO/CbzBT-B/perovskite/PI/C60/SnOx/Ag. The current density-voltage (J-V) characteristics for unmodified and buried-interface-modified devices are presented in Fig. 5a. The modified (target) device showed a marked PCE increase from 18.38 to 20.23%, primarily due to the enhanced VOC from 1.324 V to 1.382 V (Fig. 5a). The modified devices exhibited a steady‑state efficiency of 20.07%, in agreement with the J–V‑derived PCE (Supplementary Fig. 20). Negligible hysteresis and strong reproducibility were observed, with an average PCE of 19.88% (Supplementary Fig. 20). Furthermore, the 1-cm2 device achieves a PCE of 19.52% (Supplementary Fig. 21), showing good scalability of our strategy. Meanwhile, the short-circuit current density (JSC) integrated from the external quantum efficiency (EQE) matched with the J–V-extracted value, confirming the reliability of the results. By differentiating the EQE spectrum, the bandgap of champion device was 1.79 eV, implying a photovoltage loss (i.e., Eg – qVOC) of around 400 mV (Supplementary Fig. 22). The enhanced VOC stems from the reduced non-radiative recombination loss, as evidenced by external quantum efficiency of the electroluminescence (EQEEL) (Fig. 5b). The target PSC affords a higher EQEEL of 1.20% than 0.45% from the control device, validating the effectiveness of localized 2D/3D heterojunction in mitigating non-radiative recombination losses (Supplementary Fig. 23 and Supplementary Note 4)39.
a J–V curves of the control and target single-junction PSCs. b The electroluminescence external quantum efficiency (EQEEL) of the control and target PSC. c Long-term operational stability of the encapsulated PSCs at MPP tracking under continuous 1-sun equivalent illumination (without UV filter) in an N2-filled chamber without temperature control (100 mW cm−2, 40 ± 5 °C). The initial efficiency of the target PSC was 19.65%. d J-V curves of the devices based on various bandgaps. e The analysis of VOC loss for the corresponding devices. f Summary of the reported VOC values for the representative PSCs with different bandgaps (the references are reported in Supplementary Table 1). g Schematic of the structure of the monolithic (i.e., two-terminal) PO-TSC with an ICL of SnOx/Au/MoOx, and its corresponding cross-sectional SEM image. Scale bar, 300 nm. h J-V curves of the champion PSC, organic solar cell (OSC) and PO-TSC. i EQE spectra of the perovskite and organic subcell in the champion PO-TSC.
Electroluminescence (EL) was mapped to evaluate the quality of perovskite at the device level, where target devices exhibited homogeneous and brighter EL under 1.4 V bias voltage, indicating significantly suppressed non-radiative recombination at the buried interface that contributes to enhanced VOC and FF (Supplementary Fig. 24)40. Urbach energy (Eu) was determined using Fourier-transform photocurrent spectroscopy coupled with external quantum efficiency (FTPS-EQE) to assess energy disorder arising from sub-bandgap defects and lattice vibrations (Supplementary Fig. 25; Supplementary Note 5). A reduced Eu in the modified device signifies enhanced crystalline order and lattice stability attributable to more balanced crystallization processes23,33. Complementary Mott–Schottky measurements (Supplementary Fig. 26) indicate a larger built-in potential (Vbi) for the target device, which contributes to the observed increase in VOC.
We further assessed the long-term operational stability of the single-junction device under continuous one-sun illumination at the maximum power point (MPP). The target cell retained ~95% of its initial PCE after 700 h of operation (Fig. 5c). By fitting the degradation data with a biexponential decay model (R2 > 0.95), we extrapolated a projected T90 exceeding 1800 h (Fig. 5c and Supplementary Note 6)41. To assess the thermal robustness of the buried 2D/3D interface, devices were subjected to ISOS-T-1 cyclic thermal testing (25 °C ↔ 85 °C) in a nitrogen atmosphere for 600 cycles. After cycling, the devices preserved more than 94% of their initial power conversion efficiency, indicating that the buried interfacial engineering confers excellent thermal-cycling durability (Supplementary Fig. 27). This impressive device stability should be collectively attributed to the reduced trap states42, uniform halide distribution43, and suppressed ion migration (using an atomic-layer-deposited SnOx buffer layer)44. To assess the generality of the strategy, we implemented the approach on perovskite films with varying bandgap energies. (Fig. 5d and Supplementary Fig. 22). As expected, the PCEs of these cells with the localized 2D/3D heterojunction showed notable improvements, primarily due to suppressed photovoltage deficit, as summarized in Fig. 5e and Supplementary Table 1. The VOC of 1.304, 1.382, and 1.418 V achieved in this study set the record values for the reported PSCs with Eg of 1.68, 1.79, and 1.85 eV, respectively, all exceeding 90% of their thermodynamic limits. (Fig. 5e–f, Supplementary Tables 1 and 3). As summarized in Supplementary Table 4, we also deposited various ligands ITO/CbzBT-B and all the ligands enabled notably improved device performance. These results demonstrate the good versatility of our approach.
Given the minimal energy loss and impressive operational stability of the optimized WBG PSCs, they were further integrated with NBG organic sub-cells to construct monolithic PO-TSCs2,45,46,47,48. A near-infrared non-fullerene acceptor, named S9SBO-F, with an Eg of ~1.26 eV (Supplementary Fig. 28) was blended with PM6 to form a bulk-heterojunction layer in the NBG rear subcell49. Figure 5g shows the PO-TSC device architecture with SnOx/Au/MoOx serving as the interconnecting layer (ICL)50. The perovskite and organic BHJ layers were optimized to about 290 nm and 150 nm, respectively, to maximize light absorption and balance current between the subcells. The resultant tandem cell achieved a PCE of 27.11% with a high VOC of 2.169 V, FF of 83.15%, and JSC of 15.03 mA cm−2. It also exhibited negligible hysteresis and delivered a stabilized power output (SPO) of 26.9% (Fig. 5h and Supplementary Fig. 29). EQE integration yielded matched subcell currents of 15.23 mA cm⁻² for the perovskite and 15.12 mA cm⁻² for the organic cell, indicating good current matching (Fig. 5i). Across 15 individual tandem cells fabricated in different batches, the mean power conversion efficiency was 26.78% (Supplementary Fig. 26). The champion PO-TSC achieved 27.11% PCE, with a certified value of 26.3% from the National PV Industry Metrology and Testing Center (NPVM) (Supplementary Fig. 30), representing one of the highest efficiencies reported for PO-TSCs (Supplementary Fig. 31 and Supplementary Table 2). The 1-cm2 target PO-TSC delivered a remarkable PCE of 26.17%, further demonstrating the scalability of our interfacial strategy and underscoring the importance of the buried interface (Supplementary Fig. 21).
We have developed an effective and generally applicable strategy to concurrently enhance the photovoltage and realize notable operational stability in WBG PSCs. A hole-selective CbzBT-B SAM molecule obtained through rational molecular design, exhibiting improved stability and good energy alignment with WBG perovskites. It was used to induce the formation of localized 2D/3D perovskite heterojunctions at the buried interface via intricate Lewis acid-base interactions between the S heteroatoms on CbzBT-B and the ammonium groups on organic ligands. This interaction not only helps modulate perovskite crystallization to result in reduced defect density but also enhances the interfacial charge extraction. The resulting 1.79 eV-PSC showed a record VOC of 1.382 V, alongside impressive device operational stability, maintaining 95% of its initial PCE after 700 h of continuous MPP tracking. By integrating the optimized WBG PSC with an organic rear subcell, the PO-TSCs achieved a high PCE of 27.11% with a certified value of 26.3%.
Formamidinium iodide (FAI), Cesium iodide (CsI) and formamidinium bromide (FABr) were bought from Dysol. Lead bromide (PbBr2, 99.9%), Lead iodide (PbI2, 99.999%) and 1,4-diiodobenzene (DIB) were purchased from TCI. Oleylammonium iodide (OAmI), 1-Butanaminium iodide (BAI), p-Trifluorophenylethylammoniumiodide (CF3-PEAI), Lead chloride (PbCl2), fullerene (C60), methylammonium chloride (MACl) and BCP (99.9%) were supplied by Xi’an Yuri solar Co., Ltd. PEDOT:PSS 4083 (Heraeus) was bought from Advanced Electron Technology Co., Ltd. DMF (99.99%), DMSO (99.50%), IPA (99.50%), and chlorobenzene (CB, 99.90%) were obtained from J&K. PM6 was purchased from Solarmer Materials Inc. S9SBO‑F, JJ36, and PDIN‑OH were synthesized in our laboratory. Molybdenum oxide (MoOx), chloroform (CF, 99.90%), methanol (MeOH, 99.90%), and 4‑guanidinobenzoic acid hydrochloride (GBAC, 99%) were bought from Sigma‑Aldrich. High‑purity silver and gold for thermal evaporation were supplied by ZhongNuo Advanced Material (Beijing) Technology Co. The (7H‑dibenzo[c,g]carbazole) phosphonic acid (CbzNaph) and piperazinium iodide (PI) were prepared following procedures reported in our previous studies.
To prepare 1.79 eV wide-bandgap perovskite precursor (Cs0.2FA0.8Pb(I0.6Br0.4)3) with a concentration of 1.2 M, CsI (0.144 M), FAI (0.576 M), PbI2 (0.756 M), CsBr (0.096 M), FABr (0.384 M) and PbBr2 (0.48 M) were dissolved in 1 ml of mixed DMF/DMSO solvent (vol/vol, 4:1). Add 3.0 mol% of MAPbCl3 and 0.5 mol% of the GBAC to the perovskite precursor6,48. Stir the solution overnight at room temperature (RT) and use it without filtration. For 1.68 and 1.85 eV wide-bandgap perovskite precursor, keep the same procedure but adjust the I/Br ratio of (Cs0.2FA0.8Pb(IxBry)3) from 0.6/0.4 to 0.8/0.2 and 0.5/0.5, respectively.
To prepare 2D perovsikte film of (OAmI)2An-1Pbn(I0.6Br0.4)3n+1, 0.3 M precursor contain: for n = 1: OAmI (0.6 M), PbI2 (0.06 M), PbBr2 (0.24 M); for n = 2: OAmI (0.3 M), CsI (0.03 M), FAI (0.12 M), PbI2 (0.09 M), PbBr2 (0.21 M); for n = 3: OAmI (0.2 M), CsI (0.04 M), FAI (0.16 M), PbI2 (0.1 M), PbBr2 (0.2 M) were dissolved in 1 ml of mixed DMF/DMSO solvent (vol/vol, 4:1).
To formulate the BHJ precursor, PM6 and S9SBO-F were mixed at a 1:1.4 weight ratio to yield a total concentration of 14.6 mg·mL−1 in chloroform. Solid DIB was incorporated at 40% of the total donor–acceptor mass, and the solution was maintained at 60 °C with stirring for 3 h.
Prepatterned ITO glass were cleaned. After drying, substrates were baked at 100 °C for 24 h and then exposed to oxygen plasma at 50 W for 30 min. A hole‑selective SAM in IPA solution (0.5 mg·mL-1, preheated to 60 °C) was spin‑coated at 3000 rpm (ramp rate 2000 rpm·s−1) for 25 s, followed by thermal annealing for 15 min at 110 °C. For the target devices, substrates were subsequently rinsed with a ligand solution (2 mg·mL−1 in DMF) for 25 s at 3000 rpm, annealed at 100 °C for 5 min, and washed again with DMF at 3000 rpm for 25 s. Perovskite films were fabricated by one‑step spin-coating onto the SAM. 50 μL precursor was dispensed and spun at 500 rpm for 4 s, then accelerated to 4000–6500 rpm (ramp rate 1500 rpm·s−1) for 30 s. During spinning, 200 μL of CB anti‑solvent was dropped onto the perovskite film 10 s before the end of the program. Films were annealed for 10 min at 100 °C. A PI layer (0.3 mg·mL−1 in IPA) was dynamically spin‑coated for 30 s at 5000 rpm and annealed for 10 min at 100 °C. All spin‑coating steps were performed in an N₂ glovebox (O2 and H2O < 5 ppm) at ~20 °C. Electron‑transport and electrode layers were deposited by thermal evaporation in a high-vacuum environment (<5 × 10−6 torr) using a metal shadow mask (aperture area of 0.0644 cm2): 25 nm C60, 6 nm BCP (or 20 nm ALD SnOx), and 100 nm Ag. A 100 nm MgF2 antireflection coating was then evaporated onto the glass side. For J–V characterization, a non‑refractive mask defined the active area as 0.0425 cm2.
Oxygen plasma treatment (50 W, 30 min) was applied to pre-cleaned ITO substrates prior to deposition. A PEDOT:PSS film (~20 nm) was spin-coated and annealed at 150 °C for 10 min. The BHJ solution, PM6:S9SBO-F (1:1.4 w/w) at a total concentration of 14.6 mg·mL⁻¹ in chloroform, was prepared with DIB added at 40% of the combined polymer and acceptor mass; this mixture was stirred at 60 °C for 3 h before use. The blend was spin-coated onto the PEDOT:PSS layer at ambient temperature to produce an active layer of approximately 110 nm, followed by thermal annealing at 85 °C for 7 min. After cooling, a ~ 10 nm PDIN-OH interlayer (1.5 mg·mL-1 in 2,2,2-trifluoroethanol) was applied by spin coating. Devices were completed by thermal evaporation of 100 nm Ag under 5 × 10−6 torr through a 0.043 cm2 shadow mask, and the effective device area for J–V measurements was defined as 0.0324 cm2 using a non-refractive mask. Oxygen plasma treatment (50 W, 30 min) was applied to pre-cleaned ITO substrates prior to deposition. A PEDOT:PSS film (~20 nm) was spin-coated and annealed at 150 °C for 10 min. The BHJ solution, PM6:S9SBO-F (1:1.4 w/w) at a total concentration of 14.6 mg·mL-1 in chloroform, was prepared with DIB added at 40% of the combined polymer and acceptor mass; this mixture was stirred at 60 °C for 3 h before use. The blend was spin-coated onto the PEDOT:PSS layer at ambient temperature to produce an active layer of approximately 110 nm, followed by thermal annealing at 85 °C for 7 min. After cooling, a ~ 10 nm PDIN-OH interlayer (1.5 mg·mL-1 in 2,2,2-trifluoroethanol) was applied by spin coating. Devices were completed by thermal evaporation of 100 nm Ag under 5 × 10−6 torr through a 0.043 cm2 shadow mask, and the effective device area for J–V measurements was defined as 0.0324 cm2 using a non-refractive mask.
In the PO‑TSCs, the narrow‑bandgap organic subcells were deposited atop the perovskite subcells. Following the deposition of a 20 nm ALD SnOx on the wide‑bandgap subcells, a 1 nm Au was thermally evaporated, immediately followed by thermal evaporation of 10 nm MoOx onto the SnOx. A PM6:S9SBO‑F blend (1:1.5 w/w, 18 mg·mL−1 total) in chloroform with DIB additive was spin‑coated onto the MoOx layer and annealed at 85 °C for 7 min. Once cooled, a ~ 10 nm PDIN‑OH layer (1.5 mg·mL−1 in 2,2,2‑trifluoroethanol) was spin‑coated over the active film. Finally, a 100‑nm silver electrode was deposited by thermal evaporation through a metal shadow mask with an aperture area of 6.44 mm2. During J–V characterization, a non‑refractive mask defined the effective device area as 0.0425 cm2.
1H NMR spectra were obtained using Bruker AVANCE III instruments operating at 300-MHz and 400-MHz. Solution UV–vis absorption measurements were performed on an Agilent 8454 spectrophotometer. Electrochemical characterization by cyclic voltammetry was conducted with a CHI660D workstation. Transmittance and absorption spectra were acquired using a PE Lambda 750 UV–vis spectrometer. Time‑resolved photoluminescence (tdPL) was collected on a home‑built setup: a 450 nm excitation laser delivered via fiber and emission detected with an Ocean Optics USB2000 spectrometer. XPS measurements were carried out on a Thermo Fisher ESCALAB XI+ using non‑monochromatic He I (21.2 eV). Surface morphology and device cross‑sectional micrographs were acquired with a QUATTRO S scanning electron microscope. Thin‑film X‑ray diffraction measurements were performed on a D2 Phaser diffractometer employing Cu Kα radiation with a wavelength of 1.5418 Å. Time‑resolved PL measurements employed a custom setup with a 450 nm excitation delivered through an optical fiber and emission detection via an Ocean Optics USB2000 spectrometer. In‑situ PL used a 315 nm excitation under the same detection scheme. Grazing‑incidence wide‑angle X‑ray scattering experiments were performed at the BL14B1beamline of the Shanghai Synchrotron Radiation Facility (SSRF) with X‑rays of 12.398 Å (10 keV); 2D GIWAXS patterns were recorded on a MarCCD 225 detector at a 522.052 mm sample‑to‑detector distance. J–V characteristics were recorded inside an N2 glovebox with a Keithley 2400 source meter under simulated AM 1.5 G illumination (100 mW cm−2) from an SS‑F5 solar simulator, calibrated against an NREL‑certified silicon reference cell (KG‑2 filter). Active devices were masked to 0.0425 cm2; scans were executed in both directions at 10 mV s−1 with 0.02 V increments. External quantum efficiency (EQE) curves were obtained using an EQE measurement system (QE-R, EnliTech). For stability evaluation, devices with the architecture of glass/ITO/CbzBT‑B/ligands/perovskite/CF3‑PEAI/C60/SnOx/Ag were prepared using the same fabrication sequence as the champion cells. Each device was encapsulated by glass flips using a UV‑curable adhesive (LT‑U001, Luminescence Technology Corp.) and cured under 532 nm UV irradiation. Encapsulated modules were periodically characterized on a commercial MPPT platform (91PVK Solar Co. Ltd, China) mounted in a bespoke test fixture supplied with N2. A calibrated white‑light LED provided 1‑Sun equivalent illumination; MPPT measurements were conducted in air. The test chamber temperature was maintained at 45 ± 5 °C, and the system automatically calculated and applied the bias corresponding to the maximum power point.
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data supporting this study are available within the Article and Supplementary Information. Additional data are available from the corresponding author on request.
Brinkmann, K. O. et al. Perovskite–organic tandem solar cells. Nat. Rev. Mater. 9, 202–217 (2024).
Article ADS CAS Google Scholar
Jiang, X. et al. Isomeric diammonium passivation for perovskite-organic tandem solar cells. Nature 635, 860–866 (2024).
Article ADS PubMed Google Scholar
Wang, Y. et al. Homogenized contact in all-perovskite tandems using tailored 2D perovskite. Nature 635, 867–873 (2024).
Article ADS PubMed Google Scholar
Ramadan, A. J., Oliver, R. D. J., Johnston, M. B. & Snaith, H. J. Methylammonium-free wide-bandgap metal halide perovskites for tandem photovoltaics. Nat. Rev. Mater. 8, 822–838 (2023).
Article ADS CAS Google Scholar
Zhang, X. et al. Advances in inverted perovskite solar cells. Nat. Photon. 18, 1243–1253 (2024).
Zhou, Y., Herz, L. M., Jen, A. K.-Y. & Saliba, M. Advances and challenges in understanding the microscopic structure-property-performance relationship in perovskite solar cells. Nat. Energy 7, 794–807 (2022).
Article ADS CAS Google Scholar
Wu, X., Li, B., Zhu, Z., Chueh, C.-C. & Jen, A. K.-Y. Designs from single junctions, heterojunctions to multijunctions for high-performance perovskite solar cells. Chem. Soc. Rev. 50, 13090–13128 (2021).
Article CAS PubMed Google Scholar
Mahesh, S. et al. Revealing the origin of voltage loss in mixed-halide perovskite solar cells. Energy Environ. Sci. 13, 258–267 (2020).
Article CAS Google Scholar
Caprioglio, P. et al. Open-circuit and short-circuit loss management in wide-gap perovskite p-i-n solar cells. Nat. Commun. 14, 932 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhou, J. et al. Highly efficient and stable perovskite solar cells via a multifunctional hole transporting material. Joule 8, 1–16 (2024).
Article ADS CAS Google Scholar
Zhu, J. et al. A donor-acceptor-type hole-selective contact reducing non-radiative recombination losses in both subcells towards efficient all-perovskite tandems. Nat. Energy 8, 714–724 (2023).
Article ADS CAS Google Scholar
Liu, C. et al. Bimolecularly passivated interface enables efficient and stable inverted perovskite solar cells. Science 382, 810–815 (2023).
Article ADS CAS PubMed Google Scholar
Wen, J. et al. Heterojunction formed via 3D-to-2D perovskite conversion for photostable wide-bandgap perovskite solar cells. Nat. Commun. 14, 7118 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Yang, Y. et al. Amidination of ligands for chemical and field-effect passivation stabilizes perovskite solar cells. Science 386, 898–902 (2024).
Article ADS CAS PubMed Google Scholar
Xiao, Y., Yang, X., Zhu, R. & Snaith, H. J. Unlocking interfaces in photovoltaics. Science 384, 846–848 (2024).
Article ADS CAS PubMed Google Scholar
Li, X., Hoffman, J. M. & Kanatzidis, M. G. The 2D halide perovskite rulebook: how the spacer influences everything from the structure to optoelectronic device efficiency. Chem. Rev. 121, 2230–2291 (2021).
Article CAS PubMed Google Scholar
Yu, D. et al. Direct observation of photoinduced carrier blocking in mixed-dimensional 2D/3D perovskites and the origin. Nat. Commun. 13, 1–11 (2022).
Article ADS Google Scholar
Straus, D. B. & Kagan, C. R. Electrons, excitons, and phonons in two-dimensional hybrid perovskites: connecting structural, optical, and electronic properties. J. Phys. Chem. Lett. 9, 1434–1447 (2018).
Article CAS PubMed 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. 61, e202213560 (2022).
Article CAS Google Scholar
Jiang, W. et al. Spin-coated and vacuum-processed hole-extracting self-assembled multilayers with H-aggregation for high-performance inverted perovskite solar cells. Angew. Chem. Int. Ed. 63, e202411730 (2024).
Article CAS Google Scholar
Jiang, W. et al. Rational molecular design of multifunctional self-assembled monolayers for efficient hole selection and buried interface passivation in inverted perovskite solar cells. Chem. Sci. 15, 2778–2785 (2024).
Article CAS PubMed PubMed Central Google Scholar
Qu, G. et al. Conjugated linker-boosted self-assembled monolayer molecule for inverted perovskite solar cells. Joule 8, 2123–2134 (2024).
Article CAS Google Scholar
Wu, S. et al. Modulation of defects and interfaces through alkylammonium interlayer for efficient inverted perovskite solar cells. Joule 4, 1248–1262 (2020).
Article CAS Google Scholar
Mahmud, M. A. et al. Cation-diffusion-based simultaneous bulk and surface passivations for high bandgap inverted perovskite solar cell producing record fill factor and efficiency. Adv. Energy Mater. 12, 2201672 (2022).
Article CAS Google Scholar
Degani, M. et al. 23.7% efficient inverted perovskite solar cells by dual interfacial modification. Sci. Adv. 7, eabj7930 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Ferretti, A. & Prampolini, G. Complexes of alkaline and ammonium cations with dopamine and eumelanin precursors: dissecting the role of noncovalent cation−π and cation–lone pair (σ-Type) interactions. J. Phys. Chem. A 126, 2330–2341 (2022).
Article CAS PubMed Google Scholar
Laurita, G. & Seshadri, R. Chemistry, structure, and function of lone pairs in extended solids. Acc. Chem. Res. 55, 1004–1014 (2022).
Article CAS PubMed Google Scholar
Li, H. et al. 2D/3D heterojunction engineering at the buried interface towards high-performance inverted methylammonium-free perovskite solar cells. Nat. Energy 8, 946–955 (2023).
Article ADS CAS Google Scholar
Chen, M. et al. Crystal growth regulation of Ruddlesden-Popper perovskites via self-assembly of semiconductor spacers for efficient solar cells. Angew. Chem. Int. Ed. 63, e202315943 (2024).
Article CAS 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
Liu, Z. et al. All-perovskite tandem solar cells achieving >29% efficiency with improved (100) orientation in wide-bandgap perovskites. Nat. Mater. 24, 252–259 (2025).
Article ADS CAS PubMed Google Scholar
Huang, T. et al. Performance-limiting formation dynamics in mixed-halide perovskites. Sci. Adv. 7, eabj1799 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Chen, M. et al. Regulating the crystallization of mixed-halide perovskites by cation alloying for perovskite-organic tandem solar cells. Energy Environ. Sci. (2024) 17, 9580–9589 (2024).
Article CAS Google Scholar
An, Y. et al. Optimizing crystallization in wide-bandgap mixed halide perovskites for high-efficiency solar cells. Adv. Mater. 36, 2306568 (2023).
Article Google Scholar
Zheng, X. et al. Managing grains and interfaces via ligand anchoring enables 22.3%-efficiency inverted perovskite solar cells. Nat. Energy 5, 131–140 (2020).
Article ADS CAS Google Scholar
Cho, H. et al. Overcoming the electroluminescence efficiency limitations of perovskite light-emitting diodes. Science 350, 1222–1225 (2015).
Article ADS CAS PubMed Google Scholar
Wu, T. et al. Enhancing the hot carrier injection of perovskite solar cells by incorporating a molecular dipole interlayer. Adv. Funct. Mater. 32, 2204450 (2022).
Article CAS Google Scholar
El-Ballouli, A. O., Bakr, O. M. & Mohammed, O. F. Structurally tunable two-dimensional layered perovskites: from confinement and enhanced charge transport to prolonged hot carrier cooling dynamics. J. Phys. Chem. Lett. 11, 5705–5718 (2020).
Article CAS PubMed PubMed Central Google Scholar
Yang, G. et al. Stable and low-photovoltage-loss perovskite solar cells by multifunctional passivation. Nat. Photon. 15, 681–689 (2021).
Article ADS 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
Zhao, X. et al. Accelerated aging of all-inorganic, interface-stabilized perovskite solar cells. Science 377, 307–310 (2022).
Article ADS CAS PubMed Google Scholar
Torres Merino, L. V. et al. Impact of the valence band energy alignment at the hole-collecting interface on the photostability of wide band-gap perovskite solar cells. Joule 8, 585–2606 (2024).
Article Google Scholar
Wang, R. et al. Efficient wide-bandgap perovskite photovoltaics with homogeneous halogen-phase distribution. Nat. Commun. 15, 8899 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
Yang, Y. et al. Inverted perovskite solar cells with over 2000 h operational stability at 85 °C using fixed charge passivation. Nat. Energy 9, 37–46 (2024).
Article ADS CAS Google Scholar
Kang, S. et al. Boosting carrier transport in quasi-2D/3D perovskite heterojunction for high-performance perovskite/organic tandems. Adv. Mater. 37, 2411027 (2024).
Article Google Scholar
Zhang, Z. et al. Suppression of phase segregation in wide-bandgap perovskites with thiocyanate ions for perovskite/organic tandems with 25.06% efficiency. Nat. Energy 9, 592–601 (2024).
Article ADS CAS Google Scholar
Wu, X. et al. Optimization of charge extraction and interconnecting layers for highly efficient perovskite/organic tandem solar cells with high fill factor. Adv. Mater. 36, 2410692 (2024).
Article CAS Google Scholar
Wu, S. et al. Redox mediator-stabilized wide-bandgap perovskites for monolithic perovskite-organic tandem solar cells. Nat. Energy 9, 411–421 (2024).
Article ADS CAS Google Scholar
Jiang, K. et al. Suppressed recombination loss in organic photovoltaics adopting a planar–mixed heterojunction architecture. Nat. Energy 7, 1076–1086 (2022).
Article ADS Google Scholar
Chen, X. et al. Efficient and reproducible monolithic perovskite/organic tandem solar cells with low-loss interconnecting layers. Joule 4, 1594–1606 (2020).
Article CAS Google Scholar
Download references
A. K.-Y. J. thanks the sponsorship of the Lee Shau-Kee Chair Professor (Materials Science), and the support from the APRC Grant of the City University of Hong Kong (9380086, 9610508, 9610419, 9610440, and 9610492), the TCFS Grants (GHP/121/22SZ), MHKJFS Grant (MHP/054/23) and MRP Grant (MRP/040/21X) from the Innovation and Technology Commission of Hong Kong, the GRF grants (11307621, 11316422, 11308625) and CRS grants (CRS_CityU104/23, CRS_HKUST203/23) from the Research Grants Council of Hong Kong, This work was partially financially supported by City University of Hong Kong (9610739) for the project “Fostering Innovation for Resilience and Sustainable Transformation,” officially endorsed by the United Nations Educational, Scientific and Cultural Organization under the International Decade of Sciences for Sustainable Development (2024–2033). S.F.W. acknowledges the financial support from the Start-up Fund and Faculty Research Grant (SISFRG2605) from Lingnan University, the Early Career Scheme (ECS, 23300325) and the National Natural Science Foundation of China/Research Grants Council Joint Research Scheme (NSFC/RGC JRS, N_LU308/25) from the Research Grants Council of Hong Kong, and the Young Scientists Fund (Type C, 22509080) from the National Natural Science Foundation of China.
These authors contributed equally: Mingqian Chen, Wenlin Jiang, Deng Wang, Lingchen Kong.
Department of Materials Science & Engineering, City University of Hong Kong, Kowloon, Hong Kong, China
Mingqian Chen, Wenlin Jiang, Deng Wang, Lingchen Kong, Yanxun Li, Kai-kai Liu, Yunfan Wang, Qian Li, Sai-Wing Tsang & Alex K.-Y. Jen
Hong Kong Institute for Clean Energy, City University of Hong Kong, Kowloon, Hong Kong, China
Mingqian Chen, Wenlin Jiang, Deng Wang, Lingchen Kong, Kai-kai Liu, Yunfan Wang, Xiaofeng Huang & Alex K.-Y. Jen
Department of Materials Science and Engineering, and Shenzhen Engineering Research and Development Center for Flexible Solar Cells, Southern University of Science and Technology, Shenzhen, China
Deng Wang & Baomin Xu
School of Materials Science and Engineering, China University of Petroleum (East China), Qingdao, Shandong, China
Yanxun Li
Shanghai Synchrotron Radiation Facility (SSRF), Shanghai Advanced Research Institute, Chinese Academy of Sciences, Shanghai, China
Zhenhuang Su & Alex K.-Y. Jen
Department of Chemistry, City University of Hong Kong, Kowloon, Hong Kong, China
Xiaofeng Huang
Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
Xue Zheng & Jie Zhang
Wu Jieh Yee School of Interdisciplinary Studies, Lingnan University, Tuen Mun, Hong Kong, China
Xi Chen & Shengfan Wu
State Key Laboratory of Marine Environmental Health, City University of Hong Kong, Kowloon, Hong Kong, China
Alex K.-Y. Jen
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
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
PubMed Google Scholar
W.J. and S.W. conceived the idea and designed the experiments. M.C., D.W., K.L. and S.W. fabricated the devices and conducted all thin-film and device related characterizations. Y.W. helped perform the in-situ PL and was supervised by S.-W.T.; Q.L. conducted TAS test. Y.L., X.H., Y.L., X.Z., J.Z., X.C. and B.X. support the device fabrication and characterizations. W.J. contributed to the design and synthesis of hole-selective SAM. Z.S. contributed to GIWAXS measurements. M.C. wrote the first draft of the manuscript. M.C., S.W., W.J. and A.K.-Y.J. revised the manuscript. S.W. and A.K.-Y.J. supervised the project. All authors discussed the results and reviewed the manuscript.
Correspondence to Shengfan Wu or Alex K.-Y. Jen.
The authors declare no competing interests.
Nature Communications thanks Yang Yang 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
Chen, M., Jiang, W., Wang, D. et al. Localized 2D/3D heterojunction enhances photovoltage for perovskite-organic tandem solar cells. Nat Commun 17, 2093 (2026). https://doi.org/10.1038/s41467-026-68904-4
Download citation
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41467-026-68904-4
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.
