Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 May 20;22(39):e73891. doi: 10.1002/smll.73891

Synergistic Dual‐Passivation of Grain Boundaries and Buried Interface for High‐Efficiency and Stable Perovskite Solar Cells

Jiaxing Gao 1, Lijie Ling 1, Cheng Wang 1, Luyao Zhao 1, Weicun Chu 1, Luyao Li 2, Yiming Dai 1, Qiankai Ba 3, Riming Nie 1,✉, Wanlin Guo 1,✉
PMCID: PMC13360871  PMID: 42160028

ABSTRACT

The interlayer contact issue in perovskite devices is a primary source of defect generation and a major constraint on grain and interface optimization. While promoting high‐quality crystal growth and repairing interface defects are both viable pathways, these processes are often addressed separately. Here, we propose that a synergistic strategy, employing dual additives, can concurrently fulfill both objectives. Here, we report a synergistic dual‐passivation approach combining N, N‐dimethyldipropyltriamine (DMDPTF) treatment of the perovskite with cobalt acetate (Co(OAc)2) modification of the SnO2 electron transport layer. This dual passivation not only mitigates interfacial recombination but also induces an enhanced crystallization to form an optimized interlayer contact. The resulting devices achieved power conversion efficiencies (PCEs) of 25.41% (0.096 cm2 n‐i‐p), with p‐i‐n devices reaching 26.41% (0.096 cm2) and 22.18% (10.04 cm2 module). Unencapsulated devices exhibit exceptional stability, retaining 99% of initial performance for 2448 h under dark storage.

Keywords: buried interface, chemical passivation, field‐effect passivation, perovskite, stability

We present a synergistic dual‐passivation strategy combining DMDPTF treatment of the perovskite and Co(OAc)2 modification of SnO2. DMDPTF promotes crystallization and repairing grain boundaries, while Co(OAc)2 regulates bottom‐interface nucleation. The target devices exhibited a champion PCE of 25.41% for n‐i‐p cell, 26.41% for p‐i‐n cell, and retained 99% of the initial performance (T99) for 2448 h, presenting one of the most stable perovskite solar cells reported to date.

graphic file with name SMLL-22-e73891-g002.jpg

1. Introduction

Perovskite solar cells (PSCs) have made significant strides in recent years, with certified stabilized PCEs reaching up to 27.3% [1], which can be attributed to much effort in composition engineering, interfacial engineering, and theoretical modelling and simulation [2, 3, 4, 5, 6, 7, 8, 9, 10]. However, further improvements remain limited by nonradiative recombination losses, particularly at the interfaces between the perovskite absorber and transport layers (top and buried interfaces) [11, 12, 13, 14, 15]. To reduce the defects at the top interface, surface passivation strategies have been extensively explored [16, 17, 18, 19, 20, 21, 22], including the use of organohalides [23, 24, 25, 26], Lewis bases [27, 28], and dipolar compounds [29, 30]. The surface passivation strategy involves modifying defects after their formation in perovskites, and engineering nucleation and growth to reduce defects during crystallization is another parallel approach.

To address the interface issues, several excellent strategies have been proposed, such as interfacial passivation, chemical modification of the charge transport layer, and the introduction of intermediate layers to improve electronic alignment and adhesion [17, 18, 31, 32, 33, 34, 35, 36, 37, 38]. In addition to these commonly employed strategies and the defect issues that have drawn significant attention, the grain boundaries (GBs), grain surface concavities (GSCs), and grain‐boundary grooves (GBGs) formed within perovskite also influence the quality of the film [39]. More importantly, these factors have a profound negative impact on the carrier extraction at the perovskite heterojunction interface and the stability of the thin film itself, making it crucial to repair them. Clearly, the induced growth of high‐quality perovskite crystals and interface defect repair are not mutually exclusive processes [40, 41]. While a single additive often falls short of achieving both outcomes, the synergistic use of two additives provides a promising solution.

In this work, we introduce the N, N‐dimethyldipropyltriamine (DMDPTF) into the perovskite precursor solution, and cobalt acetate (Co(OAc)2) is employed to modify the buried interface between the perovskite and SnO2. DMDPTF is employed during film growth to regulate perovskite crystallization, promoting the combination of FAI and lead iodide to weaken the formation of GBs. In parallel, Co(OAc)2 is introduced at the SnO2/perovskite contact to modulate nucleation, producing more regular grain impingement and a smoother, defect‐suppressed buried interface. The devices with both additives (target devices) achieve power conversion efficiencies (PCEs) of 25.41% (0.096 cm2, n‐i‐p) and 22.18% (10.04 cm2 module), with p‐i‐n devices reaching 26.41% (0.096 cm2). The unencapsulated devices retain 99% of their initial PCE after 2448 h of storage in air at room temperature. Moreover, the devices exhibit outstanding operational stability, retaining 94.1% of the initial PCE after 1000 h under thermal stress at 60 °C and 93.6% of the initial PCE after 1200 h under maximum power point tracking (MPPT). These results position our devices among the most efficient and stable PSCs fabricated under ambient conditions to date.

2. Results and discussion

2.1. Theoretical Design of Grain Boundary Repairing

Figure 1a,b depicts several typical structural defects that may arise in polycrystalline perovskite films. Due to the heterogeneous nature of grain growth during crystallization, the perovskite layer can contain GBs, GSC, and GBGs. GBs refer to gaps where grains are not in complete contact with each other, which arise from incomplete contact between adjacent grains; GSCs are associated with interfacial depressions generated by tensile strain during grain coalescence; and GBGs form when grain corners cannot fully adapt to the substrate surface [39]. In addition, during charge transport, grain‐related defects and microscopic structural imperfections introduce nonradiative and interfacial recombination pathways, which further limit the efficiency and stability of PSCs. To address these issues, cobalt acetate is introduced at the buried interface to form acetate–lead complexes with PbI2 [42, 43, 44, 45], improve interfacial contact, and strengthen interactions between SnO2 and FAI, thereby refining nucleation at the bottom. In parallel, DMDPTF is employed to interact with FAI and promote perovskite growth, yielding a more closely packed morphology with enlarged grains. These two additives do not undergo a direct chemical reaction. Instead, they act independently yet concurrently on different regions of the perovskite film, with DMDPTF mainly influencing the bulk crystallization process and Co(OAc)2 primarily regulating the interfacial growth behavior. Although their effects are not strictly cooperative at the molecular level, their simultaneous presence effectively promotes perovskite crystallization and grain growth, leading to improved film quality and device performance.

FIGURE 1.

FIGURE 1

Open in a new tab

(a) Microscopic schematic diagram of a crystal grain. (b) Microstructure diagram of defect and modification mechanism. (c) NMR spectra of control samples. (d) NMR spectra of target samples. (e) DFT calculation of SnO2 and Co(OAc)2. (f) Defect formation energies among SnO2, Co(OAc)2, and other components. (g) DFT calculation of PVK and DMDPTF. (h) Defect formation energies in the perovskite with DMDPTF. (i) AFM spectrum of the bottom surface of the control sample. (j) Two‐dimensional height profile of the bottom surface for the control sample. (k) AFM spectrum of the bottom surface of the target sample. (l) Two‐dimensional height profile of the bottom surface for the target sample.

To further demonstrate the interaction between DMDPTF and perovskite precursor, the nuclear magnetic resonance (NMR) testing (Figure 1c,d) was conducted. In an isolated and ideally symmetric FA species, the two –NH2 groups are chemically equivalent. In the precursor/perovskite environment, however, asymmetric hydrogen‐bonding interactions with surrounding iodide species and solvent‐coordination motifs can perturb this local symmetry, resulting in nonequivalent N–H proton environments and the appearance of split peaks in the NMR spectra [46, 47]. After the introduction of DMDPTF, the peak splitting becomes less pronounced, indicating that the local asymmetric environment around FAI is weakened. This suggests that DMDPTF helps restore a more chemically equivalent local coordination environment for FAI, which is favorable for more regulated crystallization. Such interactions stabilize the precursor solution and modify the crystallization dynamics, ensuring better perovskite film quality by reducing defect formation and enhancing crystal growth. In order to improve the theoretical speculation of chemical passivation on buried interfaces, the binding energies and defect formation energies of both tin oxide and perovskite layers were calculated. Figure 1e,f present the calculated binding energies of SnO2, Co(OAc)2, and perovskite components (FAI and PbI2). After adding cobalt acetate between the SnO2 and perovskite component, the binding energy in this system was significantly enhanced. The binding energy between SnO2 and cobalt acetate is as high as 8.9690 eV. In the presence of cobalt acetate, the binding energy of the SnO2–Co(OAc)2–FAI system reaches 8.8018 eV, which is substantially higher than that of the direct SnO2–FAI interaction (3.2872 eV). A similar trend is observed for PbI2: the direct binding energy between SnO2 and PbI2 is only 2.3605 eV, whereas it increases to 3.1313 eV after the introduction of cobalt acetate. Figure 1g presents the calculated binding energies of perovskite and DMDPTF, indicating a strong binding energy between them. DMDPTF can't be incorporated into the perovskite lattice because of its molecular size and chemical nature. Instead, its role is mainly confined to the lattice surfaces [48, 49]. Figure 1h presents defect formation energies for the perovskite without and with DMDPTF. The binding energy between perovskite and DMDPTF is 1.9910 eV. After incorporating DMDPTF, the defect formation energy for iodide vacancies (VI) increased from 0.1158 to 2.6058 eV, while the defect formation energy for lead vacancies (VPb) increased from 7.3144 to 8.9888 eV, indicating that DMDPTF has a significant suppressive effect on defect formation energies. To observe the passivation results, atomic force microscope (AFM) testing and analysis were shown in Figure 1i–l. Remarkably, the target sample exhibits significantly improved grain structure at the buried interface, characterized by larger, smoother, and more uniform grains with less pronounced grain boundaries. By selecting the transverse mid‐axis to assess the uniformity of the bottom surface height, the root mean square (RMS) value of the control sample is 26.55, whereas the RMS of the target sample is only 9.00. This indicates that the buried interface of the target sample has undergone significant improvement. The AFM images of samples with individual additives (Figure S1), along with the 3D models of the target and control samples (Figures S2 and S3), also support this conclusion. Compared with the control sample, the AFM images of the DMDPTF‐treated sample show enlarged grains and a marked reduction in the clearly visible interfacial defects. Upon introducing Co(OAc)2 as a buried‐interface modifier, the buried interface becomes noticeably smoother than those of both the control and the DMDPTF‐treated sample, with particularly pronounced suppression of GSCs. The RMS roughness of the Co(OAc)2‐treated sample is lower than that of the DMDPTF‐treated sample, whereas both remain intermediate between the control and target films. Compared with the control, the 3D reconstructions of the target sample reveal a smoother morphology at both the top surface and the buried interface, together with a more uniform crystallographic orientation. Consistently, the 3D reconstructions show that the RMS values of the target sample at both the buried interface and the top surface are lower than those of the control, in agreement with the height‐profile analysis. This morphological evolution indicates that the additive successfully promotes enhanced crystallization and grain growth, resulting in a more continuous and homogeneous perovskite network. Quantitative analysis of the interfacial height distribution provides further evidence of reduced height disparities and effective grain boundary passivation.

2.2. Effects of Additives on Morphology

To further observe the passivation effect, we performed scanning electron microscope (SEM) observations on various samples. The SEM surface images presented in Figure 2a–d depict the surface morphology of the various films, respectively. With the intervention of different additives, the samples displayed larger, well‐faceted grains. Meanwhile, as the passivation strength increases, the grain boundary size becomes smaller and smaller (185 nm for the control sample, 148 nm for the DMDPTF‐treated sample, 134 nm for the Co(OAc)2‐treated sample, and 26 nm for the target sample). Specifically, the Co(OAc)2‐treated sample shows more uniform directions for GBs. Grain‐size statistics from surface SEM (Figure 2e–h) reveal a systematic rightward shift of the distribution with additive processing. The control films are dominated by sub‐micron grains (mode ≈ 0.75 µm). Introducing a single additive increases the characteristic size to ∼1.0 µm, narrows the dispersion (improved uniformity), and yields an emergent tail of > 2 µm grains. The target film exhibits a further shift to ∼1.3 µm with a markedly higher incidence of grains exceeding 2 µm, indicating additive‐enabled coarsening and a more homogeneous microstructure. For the average grain size, the average value (AVE) of the control sample is only 0.945 µm, while after adding DMDPTF and Co(OAc)2, the average value rises to 1.166 µm and 1.361 µm, and the average value of the target sample is 1.511 µm. This rise and the overall right shift of the statistical map also verified the obvious modification effect. SEM images for buried interfaces showed a similar trend to the aforementioned images (Figure 2i–l). In the control samples, residual PbI2 particulates, rough grain facets, and clearly delineated GBs, GBGs, and GSCs are evident. Introducing a single additive eliminates visible PbI2 residues and yields smoother grain surfaces; notably, DMDPTF shows optimization toward crystallization, but the repair of buried interfaces is not obvious, and Co(OAc)2 produces the more pronounced buried‐interface improvement except for the advancement of crystallization, consistent with a nucleation‐site regulation. In the target films, the buried interface is densely packed and markedly smooth, indicative of improved interfacial contact and reduced defect density. X‐ray diffraction (XRD) measurements (Figures S4 and S5) were carried out to evaluate the crystallization behavior of the films. Compared with the control, the samples treated with a single additive exhibited reduced PbI2 diffraction intensity and enhanced perovskite peaks, indicating improved phase conversion and crystallinity. Notably, the target sample showed the weakest PbI2 signal and the strongest perovskite diffraction peaks among all samples. These morphological results indicate that cobalt acetate optimizes nucleation at the buried interface, leading to a more uniform crystal orientation of perovskite and more thorough passivation of defects on the buried interface. Based on the optimization by cobalt acetate, DMDPTF further optimizes the crystallization process. The top surface remains more sensitive to factors such as solvent evaporation and local supersaturation during film formation, which can still result in a small amount of residual PbI2, even in the target sample. At the buried interface, the environment problems faced by the top interface do not exist, and Co(OAc)2 effectively regulates nucleation and interfacial contact, leading to more complete conversion of PbI2 and thus a significantly suppressed PbI2 signal. The XRD results of the buried interface reveal a small amount of residual PbI2 in the control sample, whereas this signal is markedly reduced after additive modification and becomes nearly undetectable in the target sample. Kelvin probe force microscopy (KPFM) analysis was performed on different samples. Figure 2m,n present the KPFM surface potential maps for the control and target perovskite samples. Compared to the control sample, the target sample exhibits an overall lower surface potential. This shift suggests improved electron extraction between the perovskite and the adjacent electron transport layer. The corresponding contact potential difference (CPD) maps show a homogeneous surface potential for the target sample (Figure 2o,p). The KPFM surface potential and CPD maps for the DMDPTF and Co(OAc)2 samples are shown in Figures S6 and S7. Specifically, both the surface potential and the RMS values extracted from the CPD maps decrease in the order of Control > DMDPTF > Co(OAc)2> Target, in good agreement with the other characterizations.

FIGURE 2.

FIGURE 2

Open in a new tab

(a) Interface SEM spectra for control films. (b) Interface SEM spectra for DMDPTF‐treated films. (c) Interface SEM spectra for Co(OAc)2‐treated films. (d) Interface SEM spectra for Co(OAc)2‐treated films. (e) Grain size statistical chart for control films. (f) Grain size statistical chart for DMDPTF‐treated films. (g) Grain size statistical chart for Co(OAc)2‐treated films. (h) Grain size statistical chart for target films. (i) Buried interface SEM spectra for control films. (j) Buried interface SEM spectra for DMDPTF‐treated films. (k) Buried interface SEM spectra for Co(OAc)2‐treated films. (l) Buried interface SEM spectra for Co(OAc)2‐treated films. (m) KPFM spectra for control films. (n) KPFM spectra for target films. (o) CPD map for control films. (p) CPD map for target films.

2.3. Crystallization Process Analysis

In order to investigate the changes during the crystallization process, in situ photoluminescence (PL) and in situ UV–vis (UV) tests were conducted. Figure 3a–d shows the in situ UV spectra of the control, DMDPTF‐treated, Co(OAc)2‐treated, and target samples. With increasing annealing time, the intermediate phase gradually transforms into the α‐phase, as evidenced by the absorption range extending from 400–600 nm to 400–800 nm. The phase transition times during annealing are 3.24, 3.61, 4.00, and 6.85 s for the control, DMDPTF‐ treated, Co(OAc)2‐treated, and target samples, respectively, indicating that the synergistic effect of the two additives significantly delays crystallization, providing sufficient time for perovskite molecular rearrangement and thus yielding films with enhanced crystallinity and reduced defect density [50, 51]. The crystallization kinetics and defect evolution during the annealing process were further monitored by in situ PL measurements (Figure 3e–h). Upon the individual incorporation of DMDPTF or Co(OAc)2, the PL emission intensity increased markedly, while a more pronounced enhancement was observed for the target sample, indicating that the synergistic effect of DMDPTF and Co(OAc)2 plays a pivotal role in promoting crystallization and defect passivation. The crystallinity of the perovskite films with different additives was investigated by Small‐angle X‐ray scattering (SAXS) (Figure 3i–l). In the SAXS maps, films processed with a single additive (DMDPTF or Co(OAc)2) exhibit noticeably stronger diffraction rings than the control, yet retain appreciable diffuse scattering, indicative of incomplete suppression of nanoscale disorder. By contrast, the target film shows the most intense diffraction ring and the weakest diffuse background, evidencing a highly preferred out‐of‐plane texture and superior crystallinity of the perovskite layer. Figure 3m plots the I–q profiles extracted from the four SAXS patterns, reproducing the ring‐intensity hierarchy observed in the 2D maps: the dual‐additive target film shows the strongest and sharpest features, single‐additive films are intermediate, and the control is weakest. Figure 3n shows the changes in in situ PL intensity at 800 nm over time, and Figure 3o,p show the changes in the variation of the PL curve over time, both of which are consistent with the in situ PL spectra.

FIGURE 3.

FIGURE 3

Open in a new tab

(a) In situ UV spectra for control films. (b) In situ UV spectra for DMDPTF‐treated films. (c) In situ UV spectra for Co(OAc)2‐treated films. (d) In situ UV spectra for target films. (e) In situ PL spectra for control films. (f) In situ PL spectra for DMDPTF‐treated films. (g) In situ PL spectra for Co(OAc)2‐treated films. (h) In situ PL spectra for target films. (i) SAXS spectra for control films. (j) SAXS spectra for DMDPTF‐treated films. (k) SAXS spectra for Co(OAc)2‐treated films. (l) SAXS spectra for target films. (m) The I–q profiles extracted from the four SAXS patterns. (n) The changes in in situ PL intensity at 800 nm over time for different samples. (o) The changes in the variation of the PL curve over time for the control sample. (p) The changes in the variation of the PL curve over time for the target sample.

2.4. Mechanisms of microscale Passivation

X‐ray photoelectron spectroscopy (XPS) analysis was carried out to investigate the surface state of the samples without and with DMDPTF. Compared to the control sample, the Pb 4f peaks of the DMDPTF modified film (Figure 4a) shift toward lower binding energies by ∼0.1 eV, indicating an increased electron density at the Pb2+ sites. This shift suggests that DMDPTF enhances electron donation to the lead centers, likely through coordination or electrostatic interactions. Furthermore, the decreased intensity of the Pb0 peak supports the hypothesis that DMDPTF reduces the residual PbI2 content, thereby promoting a more homogeneous perovskite phase and facilitating improved nucleation and grain growth. In addition, a distinct blue shift (∼0.8 eV) in the binding energies of both I 3d3/2 and I 3d5/2 peaks is observed in the target film, indicating a reduction in local electron density around iodine atoms, likely due to the interaction between the additives and iodide sites (Figure S8). The high‐resolution XPS spectra of N 1s in control and target perovskite films are presented in Figure S9. A discernible blue shift (∼0.73 eV) in the N 1s peak is observed in the target sample, indicating an increase in the binding energy. Figure S10 presents the XPS spectrum of Co 2p in the SnO2 layer of the target device, confirming the successful incorporation of cobalt acetate. To investigate the distribution of the two additives, buried‐interface XPS (Figure S11) measurements were conducted. The Pb and I spectra at the buried interface exhibit shifts consistent with those observed in the surface XPS results, indicating that the passivation effect of DMDPTF is also present near the buried interface. Combined with the DFT calculations and NMR results, which reveal the strong interaction between DMDPTF and FAI as well as its strong binding affinity to the perovskite components, DMDPTF is considered to be adsorbed outside the perovskite lattice, suggesting its presence throughout the perovskite film. In addition, a clear Co signal is detected at the buried interface in the Co(OAc)2‐modified sample, confirming the presence of Co species at this region and directly supporting the interfacial modification effect of cobalt acetate.

FIGURE 4.

FIGURE 4

Open in a new tab

(a) XPS spectra of PVK films. (b) UPS spectra of SnO2 films. (c) UPS spectra of PVK films. (d) PL spectra for different perovskite films. (e) TRPL spectra for different perovskite films. (f) UV absorption spectra of different PVK films. (g) FTIR spectra for different PVK films. (h) FTIR spectra for different SnO2 films.

Ultraviolet photoelectron spectroscopy (UPS) was performed to study the effect of these additives on the energy levels of the SnO2 and perovskite (Figure 4b,c). For the SnO2 layers, the valence band onset (E onset) of the Co(OAc)2‐treated sample is located 3.93 eV below the Fermi level, and the work function is 8.03 eV. The VBM and work function of the control sample are 3.86 eV and 7.88 eV (Note S1). The increase in work function indicated a stronger p‐type character induced by Co(OAc)2 incorporation. This adjustment indicates that Co(OAc)2 induces a mild p‐type behavior in SnO2, thereby improving energy level alignment with the perovskite absorber and enhancing interfacial carrier extraction. Simultaneously, the E onset of the perovskite decreased from 1.37 eV to 1.23 eV, and the work function of perovskite decreased from 5.77 eV to 5.63 eV, corresponding to KPFM results. This enhanced n‐type character favors downward band bending at the perovskite/ETL interface, facilitating electron extraction and reducing nonradiative recombination. For the energy level alignment of the buried interfaces, we conducted KPFM tests on the buried interfaces of various films (Figure S12). With the addition of additives, the buried interfaces exhibited a decrease in work function and a more uniform potential. This also demonstrates that the energy level alignment between perovskite and SnO2 on the buried interface has been improved. The energy‐level diagram and form (Figure S13 and Table S1) present the interfacial alignment before and after modification, clearly indicating that the energy‐level matching between the perovskite layer and SnO2 is markedly improved after treatment. Figure 4d shows the photoluminescence (PL) spectra of the samples with various additives. Due to the modifying effects of DMDPTF and cobalt acetate on perovskite, the PL intensity of samples with a single additive increased, indicating a reduction in non‐radiative recombination, with the target sample demonstrating the best effect. The enhanced PL intensity further confirms the suppression of nonradiative recombination pathways, contributing to improved carrier lifetimes and higher device performance. These observations were further corroborated by time‐resolved photoluminescence (TRPL) measurements (Figure 4e). The TRPL spectra showed an increase in the TRPL lifetime and enhanced carrier diffusion. The additives significantly suppressed trap‐assisted non‐radiative recombination, and the carriers were able to survive longer before recombining. Furthermore, we compared the fitted functions of the TRPL decay curves for different samples, focusing on the extracted fast (τ1) and slow (τ2) decay components after logarithmic transformation (Table S2). As the number of additives increased, both τ1 and τ2 values consistently decreased. This trend suggests a significant suppression of bulk recombination and trap‐assisted nonradiative recombination processes within the perovskite films.

The interactions between additives and perovskite were shown in UV–vis (UV) spectra and Fourier‐transform infrared (FTIR) spectra. As shown in Figure 4f, all samples exhibit a sharp absorption onset near 800 nm, consistent with the characteristic band edge of typical FAPbI3‐based perovskites. In the FTIR spectra (Figure 4g), a noticeable redshift of the C = O stretching vibration peak (near 1700 cm−1) is observed in the target sample compared to the control, indicating stronger chemical interactions or coordination between additives and perovskite components. Additionally, subtle changes in the N–H bending (∼1510 cm−1) and C–N stretching (∼1450 cm−1) regions further confirm the presence of strong interactions between DMDPTF and formamidinium (FA+) ions. The overall reduced intensity and slight shifts of these peaks in the target sample imply a decrease in free organic groups and an improvement in film compactness. The FTIR spectra for different SnO2 films (Figure S14) reveal that, compared to the control SnO2 film, the Co(OAc)2 modified film exhibits a notable decrease in the hydroxyl (Sn–OH) vibration intensity around 900–1100 cm−1, along with a slight blueshift and sharpening of the Sn–O–Sn lattice vibration near 700–800 cm−1. These observations suggest a significant reduction in surface hydroxyl groups and oxygen vacancies, as well as improved crystallinity of the SnO2 layer after modification. The enhanced lattice order and reduced trap states at the SnO2 surface are expected to facilitate more efficient electron extraction and suppress nonradiative recombination losses in the device.

2.5. Device Performance

The effect of these additives on the device based on the structure (n‐i‐p) shown in Figure 5a was also studied. The control device exhibited a short‐circuit current density (J sc) of 24.38 mA/cm2, an open‐circuit voltage (V oc) of 1.16 V, a fill factor (FF) of 82.32%, and a corresponding PCE of 23.28%. In contrast, the target device (Figure 5b) achieved a J sc of 25.63 mA/cm2, a V oc of 1.18 V, and an FF of 84.02%, resulting in a champion PCE of 25.41%. To reflect the progressiveness of this work, some previous works and their efficiencies are listed in Table S3 with this work. In addition, a target device kept a stable J sc of 24 mA/cm2 and PCE of 25.2% in the maximum power point (mpp) test for 590s (Figure 5c), demonstrating excellent operational stability. The EQE measurements (Figure 5d) further corroborate the enhanced photocurrent generation. The integrated J sc derived from the EQE spectrum increased from 24.5 mA/cm2 for the control sample to 25.04 mA/cm2 for the target device, confirming the improved charge collection efficiency. Figure S15 presents the statistical distribution of PCEs for perovskite solar cells under different additive treatments.

FIGURE 5.

FIGURE 5

Open in a new tab

(a) Schematic diagram of NIP‐PSC. (b) J–V curves for the control and target devices. (c) MPP measurement for the target device. (d) EQE spectra for the control and target devices. (e) SCLC measurement spectra for different devices. (f) Impedance spectra for different perovskite films. (g) Dark current curves for different devices. (h) Reverse scan curves of control and target cells for P‐I‐N structure. (i) Reverse and forward scan curves of the target perovskite solar module (5 cm × 5 cm).

To further investigate the carrier transport properties and trap densities in the functional layers, space‐charge‐limited current (SCLC) measurements were conducted (Figure 5e). The target sample exhibited a lower trap‐filled limit voltage (V TFL) of 0.217 V, compared to 0.257 V for the control, indicating a significant reduction in trap density after dual‐additive engineering with DMDPTF and Co(OAc)2. The progressive decrease in V TFL with the increasing number of additives reflects effective defect passivation and improved crystallinity within the perovskite bulk. The EIS impedance spectra of different devices (Figure 5f) were also measured to investigate the interface charge transfer and recombination. The target device exhibited the largest semicircle among all samples, indicating a significant increase in its recombination resistance. This result suggests that the combined use of DMDPTF and Co(OAc)2 most effectively suppresses trap‐assisted non‐radiative recombination, which is consistent with the enhanced photoluminescence/time‐resolved photoluminescence (PL/TRPL) response, reduced dark current, and improved device efficiency. The intermediate behavior exhibited by the single additive device further supports the synergistic effect of the dual‐modification strategy. An explanation of the equivalent circuit is provided in Note S2. Figure 5g displays the dark current density–voltage (J–V) curves for perovskite solar cells fabricated with different additives. Figure S16 shows the conductivity enhancement of SnO2 after the addition of Co(OAc)2. Compared to the control sample, the Co(OAc)2, DMDPTF, and the dual‐additive treated samples all show a significant reduction. This trend can be attributed to a substantial reduction in leakage currents and improved suppression of undesired carrier recombination within the perovskite films.

Motivated by the superior performance generally achieved in p‐i‐n perovskite solar cells, which is commonly attributed to the efficient hole extraction enabled by self‐assembled monolayers and the reduced non‐radiative recombination losses at the interfaces, we further extended our strategy to fabricate inverted devices using the same dual‐modification approach. For devices with p‐i‐n structure (Figure 5h), the champion target device has shown a PCE of 26.41%, while the PCE of the control device was 24.18%. We also implemented this dual‐passivation strategy in a 5 cm × 5 cm perovskite solar module (the schematic diagram of the device is shown in Figure S17), yielding a power conversion efficiency of 22.18% (Figure 5i).

2.6. Stability

Under ambient storage at room temperature (Figure 6a), the control device exhibited a rapid performance degradation, with its PCE decreasing to 80% of the initial value after approximately 500 h. In contrast, the target device maintained 99% of its initial PCE even after 2448 h of exposure. The air‐storage stability of the samples modified with an individual additive (Figure S18) positions between that of the control and the target samples. Under continuous thermal stress at 60 °C in a nitrogen‐filled glovebox (Figure 6b), the control device degraded rapidly, with its PCE dropping below 80% within 400 h. Nonetheless, the target device retained over 94.2% of its initial efficiency for more than 1000 h. Furthermore, during MPPT stability tests (Figure 6c), the PCE of the control device decreased to 80% of its initial efficiency in 600 h, whereas the target device maintained 93.6% of its original PCE after 1200 h of continuous operation. These results comprehensively demonstrate that the dual passivation strategy not only improves the initial photovoltaic performance but also significantly enhances the long‐term operational and thermal stability of perovskite solar cells under both ambient and accelerated aging conditions. In addition, thermal stability and MPPT stability of this work are compared with other works in Tables S4 and S5. These two tables show that the stabilities of our PSCs are among the highest values.

FIGURE 6.

FIGURE 6

Open in a new tab

(a) Stability data of devices stored in air. (b) Thermal stability data. (c) MPPT stability data. (d) XRD spectra of the initial device and the device after 54 h of ultraviolet irradiation for phase stability. (e) XRD spectra of the initial device and the device after 48 h of heating in air for phase stability. (f) The photo of the control film before UV irradiation. (g) The photo of the control film after 54 h UV irradiation. (h) The photo of the target film before UV irradiation. (i) The photo of the target film after 54 h UV irradiation.

In addition, to ensure the stability of the perovskite layer itself, we evaluated the phase stability. Figure 6d,e presents the phase stability evolution of different perovskite films before and after 54 h of UV exposure or 48 h of heating in air. After 63 h of continuous UV irradiation, all samples exhibited the emergence of pronounced δ‐phase peaks, a typical signature of phase degradation from the photoactive black α‐phase to the non‐photoactive yellow δ‐phase. However, the target sample maintained a substantially higher α‐phase peak intensity compared to the other samples, indicating enhanced resistance to UV‐induced phase transition. After 48 h of heating in ambient air, the control sample exhibited a significant δ‐phase peak, while there was no δ‐phase peak for the target film. These results suggest that the dual passivation strategy effectively suppresses structural degradation under UV stress, stabilizes the photoactive α‐phase, and delays the onset of detrimental phase segregation, which is crucial for improving the long‐term operational durability of perovskite solar cells. To visually highlight differences in phase stability, representative photographs are provided in Figure 6f–i. After UV exposure, the control film turns uniformly yellow across the entire substrate, indicative of extensive transition to the non‐photoactive δ phase. By contrast, the target film shows only slight yellowing at the edges while the bulk of the film remains black, demonstrating markedly superior α‐phase stability and resistance to photoinduced phase conversion.

3. Conclusions

In summary, we have developed an effective dual‐passivation strategy that synergistically combines grain boundaries and buried interface to simultaneously enhance the efficiency and stability of perovskite solar cells. By incorporating DMDPTF and Co(OAc)2 as complementary additives, we achieved a significant reduction in both bulk and interfacial defect densities. This dual modification not only promotes high‐quality perovskite crystallization but also optimizes the energy level alignment between the perovskite absorber and the SnO2 electron transport layer. Consequently, trap‐assisted nonradiative recombination is effectively suppressed, carrier extraction is enhanced, and interfacial band alignment is improved, resulting in power conversion efficiencies of 25.41% for unit cells (n‐i‐p, 0.096 cm2) and 22.18% for solar modules (10.04 cm2). For p‐i‐n devices, the PCE of the champion device is 26.41% (0.096cm2). The unencapsulated devices demonstrate outstanding operational stability, passing rigorous standard tests including dark storage, thermal aging, and MPPT. Linear extrapolation predicts T80 lifetimes of approximately 40,782 h under dark storage at room temperature, 4158 h at 60 °C thermal stress, and 3,533 h under MPPT conditions. To the best of our knowledge, this represents one of the highest performances reported to date for n‐i‐p PSCs fabricated via air‐processing, combining both exceptional efficiency and long‐term stability across multiple standard tests.

4. Experimental Section

4.1. Raw Materials

Tin (IV) oxide (15 wt% in H2O colloidal dispersion) was purchased from Alfa Aesar. PbI2 (99.99%), methylamine hydrobromide (MABr, >99.99%), CsI (99.99%), phenylethylamine Hydroiodide (PEAI, ≥98%) Spiro‐OMeTAD (99.86%), 4‐tert‐butylpyridine (96%), Li TFSI salt, dimethyl sulfoxide (DMSO, 99.9%), dimethylformamide (DMF, 99.8%), chlorobenzene (CB, 99.8%), isopropanol (IPA, 99.9%), acetonitrile (99.9%), ethanol (99.9%), MeO‐2PACz (≥98%), methyl [6,6]‐Phenyl‐C61‐butyrate (PCBM, ≥99.95%), bathocuproine (BCP, ≥99.99%), were purchased from Sigma‐Aldrich. HC(NH2)2I (formamidinium iodide (FAI), >99.99%) and methylammonium chloride (MACl, >99.99%) were purchased from GreatCell Solar. RbCl (99.95%), DMDPTF (>99%), and Co(OAc)2·4H2O (≥99.9%) were purchased from Aladdin. Nickel(II) 1‐dimethylamino‐2‐methyl‐2‐butoxide (Ni(dmamb)2, ≥99.9%) was purchased from Suzhou Origin Materials Technology Co., Ltd.

4.2. Preparation of Precursor Solutions

4.2.1. N‐I‐P Devices

The ETL precursor was prepared by diluting the SnO2 colloidal solution with deionized water (1:3 wt). For target devices, 2 mg Co(OAc)2·4H2O was added to 1 mL SnO2 colloidal solution. FAPbI3 perovskite solution was prepared by mixing PbI2 (714.55 mg), FAI (266.6 mg), RbCl (5 mg), MACl (32.41 mg) in 888 µL DMF and 112 µL DMSO. For target devices, 5 µL mixed solution (DMDPTF: DMF = 2:98) was added to every 100 µL perovskite solution. The HTL precursor solution was prepared by dissolving 72.3 mg spiro‐OMeTAD in 1 mL chlorobenzene with 35 µL Li‐TFSI solution (260 mg mL−1 in acetonitrile) and 28.8 µL 4‐tBP.

4.2.2. P‐I‐N Devices

SAM precursor was prepared by dissolving 0.5 mg MeO‐2PACz in 1 mL of ethanol. For target devices, Co(OAc)2·4H2O (2 mg mL−1 in ethanol) was added to a 1 mL SAM colloidal solution. FAPbI3 perovskite solution was prepared by mixing PbI2 (664.7 mg), FAI (216.7 mg), MAI (11.27 mg), CsI (18.17 mg), and MACl (15.9 mg) in 888 µL DMF and 112 µL DMSO. For target devices, 5 µL mixed solution (DMDPTF: DMF = 2:98) was added to every 100 µL perovskite solution. PCBM was dissolved in CB to 10 mg/mL. BCP was dissolved in IPA for 1 mg/mL.

4.3. Fabrication of Perovskite Solar Cells

4.3.1. N‐I‐P Devices

The glass/FTO substrate was sequentially washed with isopropanol, acetone, distilled water, and isopropanol. After drying, the substrates were further cleaned by ultraviolet‐ozone (UVO) treatment for 15 min before use. For ETL films, the SnO2 solution was spin‐coated on glass/ITO substrates at 3,000 rpm for 30s in ambient air, and then annealed at 150 °C for 30 min. After drying and UVO treatment (15 min), the perovskite precursor solution was spin‐coated onto the ETL at i) 1,000 rpm for 10 s, ii) 4 000 rpm for 25s, where 200 µL CB was dripped onto the film at the 15s of stage (ii) until the end. Then the film was annealed at 120°C for 60 min. PEAI (5 mg mL−1 in IPA) passivation layer was spin‐coated onto the perovskite film at 4,000 rpm for 30s. The HTL solution was coated at 3,000 rpm for 30 s. Finally, an Au back electrode was deposited by thermal evaporation at a pressure of 3 × 10−6 torr.

4.3.2. P‐I‐N Devices

NiOx layer was prepared by atomic layer deposition (ALD) for ∼20 nm after the same washing, drying, and UVO treatment as N‐I‐P devices. SAM precursor was spin‐coated on the NiOx film at 3000 rpm for 30s, and then annealed at 100°C for 10 min. The perovskite precursor solution was spin‐coated onto the HTL at i) 1000 rpm for 10 s, ii) 5000 rpm for 30s, where 200 µL CB was dripped onto the film at the 10s stage (ii) until the end. Then the film was annealed at 120 °C for 20 min. PEAI (5 mg mL−1 in IPA) passivation layer was spin‐coated onto the perovskite film at 4,000 rpm for 30s. The PCBM and BCP solution was coated at 4,000 rpm for 30 s in order. Finally, an Ag back electrode was deposited by thermal evaporation at a pressure of 3 × 10−6 torr.

4.4. Fabrication of Perovskite Modules

The FAPbI3 mini‐modules include six sub‐cells connected in series on a 50 × 50 mm2 ITO substrate. 90 µm wide P1 lines were first scribed using a laser. A 20 nm NiOx layer was prepared by the ALD method (finished by the ALD‐R900 machine from Shanghai Force Stars Technology Co, Ltd; the relevant characterization is shown in Figures S19–S21) with pumping Ni(dmamb)2 (gas), N2, O2 in O3 atmosphere under 300°C. During pumping, the Ni(dmamb)2 tank was heated at 90°C, and the rotation plane workholder kept spinning at 100 rpm. The SAM precursor was then spin‐coated as described in P‐I‐N solar cell fabrication. FAPbI3 perovskite solution was spin‐coated onto NiOx/SAM substrate at i) 1000 rpm for 5 s, ii) 4000 rpm for 18 s in ambient air. After spin coating, the film was transferred into a vacuum flash evaporator (VCD), held under vacuum for 30 s. After vacuum quenching, the intermediate films were annealed at 120°C for 30 min. PEAI, PCBM, and BCP solutions were coated sequentially as described in solar cell fabrication. Then the P2 lines were laser scribed for 50 µm wide using a picosecond laser with a wavelength of 532 nm. After P2 processing, a 100 nm thick Au electrode was deposited by thermal evaporation and then separated by P3 lines scribed using a picosecond laser with a wavelength of 532 nm. The width of the dead area was about 700 µm. The active area of a device was around 10.04 cm2.

4.5. Characterizations

The current density–voltage (J–V) curves were measured using an AM 1.5G solar simulator equipped with a Xenon lamp (USHIO) and a Keithley 2450 source meter. The light intensity was calibrated to be 100 mW/cm2 using a NIST‐certified monocrystalline Si solar cell. The PL and TRPL were performed via a 450 nm laser in FLS 1000, Edinburgh Instruments. The surface chemical environment of perovskite films was obtained by X‐ray photoelectron spectrometer with binding energy referenced to the C 1s peak at 284.8 eV (XPS, Thermo Scientific ESCALAB Xi+). UV–vis spectra were measured with a Varian Cary5. The SEM images of the perovskite films were measured by a field‐emission scanning electron microscope (FE‐SEM, Zeiss Sigma 300). And the AFM and KPFM images of the perovskite films were taken by Horiba JY Labram EVO. The Mott‐Schottky curves and EIS data were measured by an electrochemical workstation (AMETEK PARSTAT 3000‐DX). The XRD patterns were measured by BRUKER D2 PHASER. Additional information on structures and chemical bonding was measured using Fourier‐transformed infrared spectroscopy (FTIR, Thermo Scientific Nicolet iS50) and thermogravimetric analysis (TGA, NETZSCH STA 449F3). Texture properties were evaluated with the N2‐sorption technique (Micromeritics APSP 2460).

Funding

This work was supported by financial support from National Natural Science Foundation of China (52203359). This work was also supported by the Research Fund of State Key Laboratory of Mechanics and Control for Aerospace Structures (Nanjing University of Aeronautics and astronautics) (Grant No. MCAS‐I‐0525K01), the Fund of Basic Research Program of Jiangsu (BK20253025) and the Fund of Prospective Layout of Scientific Research for NUAA (Nanjing University of Aeronautics and Astronautics).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: smll73891‐sup‐0001‐SuppMat.doc.

SMLL-22-e73891-s001.doc (10.6MB, doc)

Acknowledgements

This work was supported by financial support from the National Natural Science Foundation of China (52203359). This work was also supported by the National and Jiangsu Province NSF (T2293691, BK20212008) of China, National Key Research and Development Program of China (2019YFA0705400), the Research Fund of National Key Laboratory of Mechanics and Control for Aerospace Structures (MCMS‐I‐0422K01), the Fundamental Research Funds for the Central Universities (NC2023001, NJ2023002, NJ2022002) and the Fund of Prospective Layout of Scientific Research for NUAA (Nanjing University of Aeronautics and Astronautics). We acknowledge the characterization support of the Center for Microscopy and Analysis at Nanjing University of Aeronautics and Astronautics. The research work is supported by the funds for talents of Nanjing University of Aeronautics and Astronautics.

Contributor Information

Riming Nie, Email: rmnie@nuaa.edu.cn.

Wanlin Guo, Email: wlguo@nuaa.edu.cn.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  • 1. Green M. A., Dunlop E. D., Yoshita M., et al., “Solar Cell Efficiency Tables (Version 66),” Progress in Photovoltaics: Research and Applications 33, no. 7 (2025): 795–810, 10.1002/pip.3919. [DOI] [Google Scholar]
  • 2. Park J., Kim J., Yun H. S., et al., “Controlled Growth of Perovskite Layers with Volatile Alkylammonium Chlorides,” Nature 616, no. 7958 (2023): 724–730, 10.1038/s41586-023-05825-y. [DOI] [PubMed] [Google Scholar]
  • 3. Jiang Q., Tirawat R., Kerner R. A., et al., “Towards Linking Lab and Field Lifetimes of Perovskite Solar Cells,” Nature 623, no. 7986 (2023): 313–318, 10.1038/s41586-023-06610-7. [DOI] [PubMed] [Google Scholar]
  • 4. Tan S., Huang T. Y., Yavuz I., et al., “Stability‐Limiting Heterointerfaces of Perovskite Photovoltaics,” Nature 605, no. 7909 (2022): 268–273, 10.1038/s41586-022-04604-5. [DOI] [PubMed] [Google Scholar]
  • 5. Liu S. C., Lu Y., Yu C., et al., “Triple‐Junction Solar Cells with Cyanate in Ultrawide‐Bandgap Perovskites,” Nature 628, no. 8007 (2024): 306–312, 10.1038/s41586-024-07226-1. [DOI] [PubMed] [Google Scholar]
  • 6. Meng H. G., Mao K. T., Cai F. C., et al., “Inhibition of Halide Oxidation and Deprotonation of Organic Cations with Dimethylammonium Formate for Air‐processed p–i–n Perovskite Solar Cells,” Nature Energy 9, no. 5 (2024): 536–547, 10.1038/s41560-024-01471-4. [DOI] [Google Scholar]
  • 7. Tessler N. and Vaynzof Y., “Insights from Device Modeling of Perovskite Solar Cells,” ACS Energy Letters 5, no. 4 (2020): 1260–1270, 10.1021/acsenergylett.0c00172. [DOI] [Google Scholar]
  • 8. Tang H. C., Shen Z. C., Shen Y. Z., et al., “Reinforcing Self‐assembly of Hole Transport Molecules for Stable Inverted Perovskite Solar Cells,” Science 383, no. 6688 (2024): 1236–1240, 10.1126/science.adj9602. [DOI] [PubMed] [Google Scholar]
  • 9. Gulia Y., Pai N., Satapathi S., and Angmo D., “FAPbI3 Powder Synthesis Routes for State‐of‐the‐Art Perovskite Solar Cells: A Review,” Small Structures 6 (2025): 202500414. [Google Scholar]
  • 10. Yusoff A. B., Vasilopoulou M., Georgiadou D. G., Palilis L. C., Abate A., and Nazeeruddin M. K., “Passivation and Process Engineering Approaches of Halide Perovskite Films for High Efficiency and Stability Perovskite Solar Cells,” Energy & Environmental Science 14, no. 5 (2021): 2906–2953, 10.1039/D1EE00062D. [DOI] [Google Scholar]
  • 11. Liu S. W., Biju V. P., Qi Y. B., Chen W., and Liu Z. H., “Recent Progress in the Development of High‐Efficiency Inverted Perovskite Solar Cells,” Npg Asia Materials 15, no. 27 (2023): 27. [Google Scholar]
  • 12. Munir R., Sheikh A. D., Abdelsamie M., et al., “Hybrid Perovskite Thin‐Film Photovoltaics: In Situ Diagnostics and Importance of the Precursor Solvate Phases,” Advanced Materials 29, no. 2 (2017): 1604113, 10.1002/adma.201604113. [DOI] [PubMed] [Google Scholar]
  • 13. Xiong Z., Chen X., Zhang B., et al., “Simultaneous Interfacial Modification and Crystallization Control by Biguanide Hydrochloride for Stable Perovskite Solar Cells with PCE of 24.4%,” Advanced Materials 34, no. 8 (2022): 2106118, 10.1002/adma.202106118. [DOI] [PubMed] [Google Scholar]
  • 14. Nie R., Zhang P., Gao J., et al., “Enhanced Coordination Interaction with Multi‐Site Binding Ligands for Efficient and Stable Perovskite Solar Cells,” Nature Communications 16, no. 1 (2025): 6438, 10.1038/s41467-025-61563-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Nie R., Dai Y., Wang R., et al., “Enhanced Stability and Efficiency in Perovskite Solar Cells via Mixed‐Metal Chalcohalide‐Alloyed Formamidinium Lead Iodide,” Nature Communications 16, no. 1 (2025): 7343, 10.1038/s41467-025-62125-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Dong W., Zhu C. P., Bai C., et al., “Low‐Cost Hydroxyacid Potassium Synergists as an Efficient In Situ Defect Passivator for High Performance Tin‐Oxide‐Based Perovskite Solar Cells,” Angewandte Chemie International Edition 62, no. 25 (2023): 202302507, 10.1002/anie.202302507. [DOI] [PubMed] [Google Scholar]
  • 17. Zhang B., Oh J., Sun Z., et al., “Buried Guanidinium Passivator with Favorable Binding Energy for Perovskite Solar Cells,” ACS Energy Letters 8, no. 4 (2023): 1848–1856, 10.1021/acsenergylett.2c02881. [DOI] [Google Scholar]
  • 18. Xiong Z., Chen X., Zhang B., et al., “Simultaneous Interfacial Modification and Crystallization Control by Biguanide Hydrochloride for Stable Perovskite Solar Cells with PCE of 24.4%,” Advanced Materials 34, no. 8 (2022): 2106118, 10.1002/adma.202106118. [DOI] [PubMed] [Google Scholar]
  • 19. Zhu P. C., Gu S., Luo X., et al., “Simultaneous Contact and Grain‐Boundary Passivation in Planar Perovskite Solar Cells Using SnO2 ‐KCl Composite Electron Transport Layer,” Advanced Energy Materials 10, no. 3 (2020): 1903083, 10.1002/aenm.201903083. [DOI] [Google Scholar]
  • 20. Guo J. J., Sun J. G., Hu L., et al., “Indigo: A Natural Molecular Passivator for Efficient Perovskite Solar Cells,” Advanced Energy Materials 12, no. 22 (2022): 2200537, 10.1002/aenm.202200537. [DOI] [Google Scholar]
  • 21. Xu J., Chen H., Grater L., et al., “Anion Optimization for Bifunctional Surface Passivation in Perovskite Solar Cells,” Nature Materials 22, no. 12 (2023): 1507–1514, 10.1038/s41563-023-01705-y. [DOI] [PubMed] [Google Scholar]
  • 22. Liu S. W., Li J. B., Xiao W. S., et al., “Buried Interface Molecular Hybrid for Inverted Perovskite Solar Cells,” Nature 632, no. 8025 (2024): 536–542, 10.1038/s41586-024-07723-3. [DOI] [PubMed] [Google Scholar]
  • 23. Li T. T., Xu J., Lin R. X., et al., “Inorganic Wide‐bandgap Perovskite Subcells with Dipole Bridge for all‐perovskite Tandems,” Nature Energy 8, no. 6 (2023): 610–620, 10.1038/s41560-023-01250-7. [DOI] [Google Scholar]
  • 24. Chen H., Teale S., Chen B., et al., “Quantum‐Size‐Tuned Heterostructures Enable Efficient and Stable Inverted Perovskite Solar Cells,” Nature Photonics 16, no. 5 (2022): 352–358, 10.1038/s41566-022-00985-1. [DOI] [Google Scholar]
  • 25. Lin R. X., Xu J., Wei M. Y., et al., “All‐perovskite Tandem Solar Cells with Improved Grain Surface Passivation,” Nature 603, no. 7899 (2022): 73–78, 10.1038/s41586-021-04372-8. [DOI] [PubMed] [Google Scholar]
  • 26. Kim D. H., Muzzillo C. P., Tong J. H., et al., “Bimolecular Additives Improve Wide‐Band‐Gap Perovskites for Efficient Tandem Solar Cells with CIGS,” Joule 3, no. 7 (2019): 1734–1745, 10.1016/j.joule.2019.04.012. [DOI] [Google Scholar]
  • 27. Li Z., Li B., Wu X., et al., “Organometallic‐functionalized Interfaces for Highly Efficient Inverted Perovskite Solar Cells,” Science 376, no. 6591 (2022): 416–420, 10.1126/science.abm8566. [DOI] [PubMed] [Google Scholar]
  • 28. Li X. D., Zhang W. X., Guo X. M., Lu C. Y., Wei J. Y., and Fang J. F., “Constructing Heterojunctions by Surface Sulfidation for Efficient Inverted Perovskite Solar Cells,” Science 375, no. 6579 (2022): 434–437, 10.1126/science.abl5676. [DOI] [PubMed] [Google Scholar]
  • 29. Caprioglio P., Cruz D. S., Caicedo‐Dávila S., et al., “Bi‐functional Interfaces by Poly(ionic liquid) Treatment in Efficient Pin and Nip Perovskite Solar Cells,” Energy & Environmental Science 14, no. 8 (2021): 4508–4522, 10.1039/D1EE00869B. [DOI] [Google Scholar]
  • 30. Zhu Z. J., Mao K. T., Zhang K., et al., “Correlating the Perovskite/Polymer Multi‐Mode Reactions with Deep‐Level Traps in Perovskite Solar Cells,” Joule 6, no. 12 (2022): 2849–2868, 10.1016/j.joule.2022.10.007. [DOI] [Google Scholar]
  • 31. Liu C., Yang Y., Chen H., et al., “Bimolecularly Passivated Interface Enables Efficient and Stable Inverted Perovskite Solar Cells,” Science 382, no. 6672 (2023): 810–815, 10.1126/science.adk1633. [DOI] [PubMed] [Google Scholar]
  • 32. Qi W. J., Liu Z., Xie X. R., et al., “A Graded Redox Interfacial Modifier for High‐Performance Perovskite Solar Cells,” Angewandte Chemie International Edition 63, no. 50 (2024): 202411604, 10.1002/anie.202411604. [DOI] [PubMed] [Google Scholar]
  • 33. Yang Y., Chen H., Liu C., et al., “Amidination of Ligands for Chemical and Field‐Effect Passivation Stabilizes Perovskite Solar Cells,” Science 386, no. 6724 (2024): 898–902, 10.1126/science.adr2091. [DOI] [PubMed] [Google Scholar]
  • 34. Kim C., Kim K., Kim Y., et al., “Trimming Defective Perovskite Layer Surfaces for High‐Performance Solar Cells,” Energy & Environmental Science 17, no. 22 (2024): 8582–8592, 10.1039/D4EE03585B. [DOI] [Google Scholar]
  • 35. Song Z. L., Zou Y., Gao Y. P., et al., “Buried and Bulk Synergistic Engineering Enables High‐Performance Inverted 2D/3D Perovskite Solar Cells,” Energy & Environmental Science 18, no. 8 (2025): 3740–3749, 10.1039/D5EE00156K. [DOI] [Google Scholar]
  • 36. Chen X., Wang Q., Wei H., et al., “Minimizing the Buried Interfacial Energy Loss Using a Fluorine‐Substituted Small Molecule for 25.92%‐Efficiency and Stable Inverted Perovskite Solar Cells,” Energy & Environmental Science 17, no. 19 (2024): 7342–7354, 10.1039/D4EE02964J. [DOI] [Google Scholar]
  • 37. Menzel D., Al‐Ashouri A., Tejada A., et al., “Field Effect Passivation in Perovskite Solar Cells by a LiF Interlayer,” Advanced Energy Materials 12, no. 30 (2022): 2201109, 10.1002/aenm.202201109. [DOI] [Google Scholar]
  • 38. Li L., Wang C., Chu W., et al., “Fully Chemical Interface Engineering for Statically and Dynamically Stable Perovskite Solar Cells,” Nature Communications 16, no. 1 (2025): 8575, 10.1038/s41467-025-63588-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Xiao T., Hao M., Duan T., et al., “Elimination of Grain Surface Concavities for Improved Perovskite Thin‐Film Interfaces,” Nature Energy 9, no. 8 (2024): 999–1010, 10.1038/s41560-024-01567-x. [DOI] [Google Scholar]
  • 40. Chen R., Wang J., Liu Z., et al., “Reduction of Bulk and Surface Defects in Inverted Methylammonium‐ and Bromide‐Free Formamidinium Perovskite Solar Cells,” Nature Energy 8, no. 8 (2023): 839–849, 10.1038/s41560-023-01288-7. [DOI] [Google Scholar]
  • 41. Park J. K., Heo J. H., Jeon H. S., et al., “Molecularly Tailored Dual‐Interface Passivation via Solvent‐Free Rub‐on Transfer for Efficient and Stable Perovskite LEDs,” Science Advances 11, no. 43 (2025): ady4671, 10.1126/sciadv.ady4671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Gao D. P., Li B., Li Z., et al., “Highly Efficient Flexible Perovskite Solar Cells through Pentylammonium Acetate Modification with Certified Efficiency of 23.35%,” Advanced Materials 35, no. 3 (2023): 2206387, 10.1002/adma.202206387. [DOI] [PubMed] [Google Scholar]
  • 43. Zhao J., Fürer S. O., McMeekin D. P., et al., “Efficient and Stable Formamidinium–caesium Perovskite Solar Cells and Modules from Lead Acetate‐based Precursors,” Energy & Environmental Science 16, no. 1 (2023): 138–147, 10.1039/D2EE01634F. [DOI] [Google Scholar]
  • 44. Li M. H., Zhang R. X., Guo Y., et al., “Introducing Lead Acetate into Stoichiometric Perovskite Lewis Acid‐base Precursor for Improved Solar Cell Photovoltaic Performance,” Journal of Alloys and Compounds 767 (2018): 829–837, 10.1016/j.jallcom.2018.07.190. [DOI] [Google Scholar]
  • 45. Liu R. C., Deng C. Y., Li G. D., et al., “Optimization of α‐FAPbI3 Crystallization by Intermediate Compounds Transformation for Efficient and Stable Perovskite Solar Cells,” Chemical Engineering Journal 502 (2024): 157734, 10.1016/j.cej.2024.157734. [DOI] [Google Scholar]
  • 46. Huang Y., Shi Y., Cohen T. A., Ho K., and Luscombe C. K., “Studying Hydrogen–Halide Interactions in Lead Halide Perovskite with Isoelectronic Cations,” Chemistry of Materials 35, no. 20 (2023): 8417–8425, 10.1021/acs.chemmater.3c01149. [DOI] [Google Scholar]
  • 47. Garrote‐Márquez A., Lodeiro L., Suresh R., Grau‐Crespo R., Menéndez‐Proupin E., and Cruz Hernández N., “Hydrogen Bonds in Lead Halide Perovskites: Insights from Ab Initio Molecular Dynamics,” The Journal of Physical Chemistry C 127, no. 32 (2023): 15901–15910, 10.1021/acs.jpcc.3c02376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Zhang K., Ding B., Wang C., et al., “Highly Efficient and Stable FAPbI3 Perovskite Solar Cells and Modules Based on Exposure of the (011) Facet,” Nano‐Micro Letters 15, no. 1 (2023): 138, 10.1007/s40820-023-01103-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Zhang L., Yu F., Chen L., and Li J., “Adsorption of Molecular Additive Onto Lead Halide Perovskite Surfaces: A Computational Study on Lewis Base Thiophene Additive Passivation,” Applied Surface Science 443 (2018): 176–183, 10.1016/j.apsusc.2018.02.251. [DOI] [Google Scholar]
  • 50. Han Y., Fu J., Ren Z., et al., “Inorganic Perovskite/Organic Tandem Solar Cells with 25.1% Certified Efficiency via Bottom Contact Modulation,” Nature Energy 10, no. 4 (2025): 513–525, 10.1038/s41560-025-01742-8. [DOI] [Google Scholar]
  • 51. Wang J., Wang P., Chen T., et al., “Isomerism Effect of 3D Dimeric Acceptors for Non‐Halogenated Solvent‐Processed Organic Solar Cells with 20% Efficiency,” Angewandte Chemie International Edition 64, no. 12 (2025): 202423562, 10.1002/anie.202423562. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting File: smll73891‐sup‐0001‐SuppMat.doc.

SMLL-22-e73891-s001.doc (10.6MB, doc)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Small (Weinheim an Der Bergstrasse, Germany) are provided here courtesy of Wiley

RESOURCES