I introduced 1-adamantylamine hydrochloride (ADA) as an interfacial modifier between tin dioxide (SnO2) and the perovskite layer to form an electron reflux barrier layer in perovskite solar cells. The ADA layer was deposited by spin coating from isopropanol solutions at different mass concentrations. The amino lone pair electrons in ADA can coordinate with under-coordinated Pb2+ and passivate oxygen vacancies on SnO2, which improves perovskite film quality, reduces defect density, and enhances open-circuit voltage. The optimal ADA-modified perovskite solar cells achieved a power conversion efficiency (PCE) of 23.07%, an open-circuit voltage of 1.13 V, a short-circuit current density of 24.67 mA cm−2, and a fill factor of 82.59%. After storage in air for 1000 h, the modified devices retained more than 85% of their initial PCE. Both efficiency and stability were significantly enhanced. This work demonstrates that a thin cage-like molecular layer can simultaneously passivate buried interfaces, block electron reflux, and suppress non-radiative recombination in perovskite solar cells.
Perovskite solar cells have emerged as one of the most promising photovoltaic technologies because of the excellent solution processability and outstanding optoelectronic properties of their photoactive materials. The electron transport layer (ETL) plays a critical role in extracting and transporting photogenerated electrons. However, low-temperature-processed SnO2 inevitably contains defects at grain boundaries and interfaces. These defects, together with different oxidation states of Sn and O, trigger non-radiative recombination, dynamic losses, and energy losses during transport, ultimately limiting the open-circuit voltage (VOC) and fill factor (FF) of perovskite solar cells.
During thermal annealing, SnO2 readily forms tin interstitials, oxygen vacancies, and surface hydroxyl defects. At the same time, annealing of the perovskite (PVK) layer can create voids at the buried interface because of solvent evaporation. These defects aggravate non-radiative recombination and impede carrier transport, making the deep defect density at the buried interface higher than that at the top perovskite interface. Therefore, interfacial engineering between the ETL and the perovskite layer is essential for high-performance perovskite solar cells.
To eliminate deep defects and improve the overall performance of perovskite solar cells, I used a cage-like molecule, ADA, as an interface modifier between SnO2 and the perovskite layer. Adamantane (AD) is a highly symmetric, thermodynamically stable, and photostable cage compound. It is easily soluble in non-polar organic solvents, and its derivatives have broad application prospects. I compared AD and ADA modifiers because both contain a rigid adamantane cage, while ADA additionally provides an ammonium chloride group that can interact with both SnO2 and the perovskite layer. The chloride anion can passivate oxygen vacancies on SnO2, and the amino group can coordinate with under-coordinated Pb2+ in the perovskite. This dual interaction is expected to reduce interfacial defects, improve film quality, and create an ultrathin tunneling layer that permits forward electron transport while blocking electron reflux in perovskite solar cells.
The device fabrication flow is summarized as follows. ITO conductive glass substrates were cleaned by ultrasonication in deionized water and anhydrous ethanol for 30 min each. After drying in an oven, the ITO substrates were treated with ultraviolet-ozone for 15 min. A diluted SnO2 aqueous colloid dispersion (5 wt%) was spin-coated onto the ITO substrate at 3500 r min−1 for 30 s and annealed at 150 °C for 30 min. For ADA-modified devices, I prepared isopropanol solutions of ADA at mass concentrations of 0.1, 0.3, and 0.5 mg mL−1. At least ten independent devices were fabricated for each concentration. The ADA solution was spin-coated onto the SnO2 film at 4000 r min−1 for 30 s and annealed at 100 °C for 5 min. The resulting SnO2 films had different surface ADA contents.
In an inert glovebox, I dissolved 691.5 mg PbI2 in 1 mL of a mixed DMF/DMSO solvent with a volume ratio of 9:1. Then I dissolved MAI (6.39 mg), MACl (9.00 mg), and FAI (90.00 mg) in 1 mL IPA to form a mixed amine salt solution. The PbI2 precursor solution was spin-coated onto the SnO2 film at 1500 r min−1 for 25 s and annealed at 70 °C for 1 min. The mixed amine salt solution was spin-coated onto the PbI2 film at 2000 r min−1 for 30 s, annealed at 30 °C for 5 min, and then annealed at 150 °C for 15 min to form the perovskite layer. For the hole transport layer (HTL), I dissolved 90 mg Spiro-OMeTAD in 1 mL chlorobenzene and added 23 µL Li-TFSI solution (520 mg Li-TFSI in 1 mL acetonitrile) and 39.5 µL t-BP. The HTL solution was deposited onto the perovskite film at 3000 r min−1 for 30 s. After drying for 12 h, silver electrodes were deposited by thermal vacuum evaporation to complete devices with the structure ITO/SnO2/PVK/Spiro-OMeTAD/Ag. The active area was 0.0625 cm2.
Current density–voltage (J–V) curves were measured with a 2420 source meter and an Oriel Sol3A solar simulator providing 100 mW cm−2 illumination. External quantum efficiency (EQE) spectra were obtained with a QE-R system. Scanning electron microscopy (SEM) was used to analyze the morphology and structure of perovskite films and device cross sections. Atomic force microscopy (AFM) was used to obtain surface roughness by scanning the sample surface with a sharp probe. X-ray diffraction (XRD) was performed to study crystallinity. Steady-state photoluminescence (PL) spectra were measured with an excitation wavelength of 475 nm. Time-resolved photoluminescence (TRPL) spectra were measured using time-correlated single-photon counting with a laser wavelength of 760 nm. X-ray photoelectron spectroscopy (XPS) and ultraviolet photoelectron spectroscopy (UPS) were performed to analyze chemical states and energy levels. Ultraviolet–visible absorption spectra were recorded. Electrochemical impedance spectroscopy (EIS) and Mott–Schottky analysis were carried out on an electrochemical workstation.
The photoelectric performance and stability of perovskite solar cells are strongly influenced by the buried interface between SnO2 and the perovskite layer. The molecular electrostatic potential map shows that the electron cloud density of AD and ADA is mainly distributed on the amine group. This indicates that the amine group in ADA has strong Lewis base character and can coordinate with oxygen vacancy defects in SnO2 and under-coordinated Pb2+ in the perovskite. Compared with AD, ADA has a larger steric hindrance because of the ammonium chloride group, which significantly affects adsorption on SnO2 or the perovskite surface. The interplay between coordination ability and steric hindrance determines the optimal ADA concentration.
I examined the effect of ADA concentration on device performance. When the mass concentration of ADA was 0.5 mg mL−1, the PCE was much lower than that of devices prepared with 0.3 mg mL−1. At high concentration, the ADA layer becomes thicker and significantly suppresses charge transfer. When the ADA concentration is too low, device performance also decreases because the passivation effect is insufficient. After repeated fabrication and testing, I found that devices prepared with 0.3 mg mL−1 ADA consistently achieved the highest performance. Therefore, 0.3 mg mL−1 was chosen as the optimized concentration for the SnO2-ADA electron transport layer.
The photovoltaic performance data of perovskite solar cells are summarized in Table 1. The pristine SnO2 device showed a PCE of 20.22%, a JSC of 24.52 mA cm−2, a VOC of 1.06 V, and an FF of 78.16%. Under the same measurement conditions, the SnO2-ADA device achieved a PCE of 23.07%, a JSC of 24.67 mA cm−2, a VOC of 1.13 V, and an FF of 82.59%. The SnO2-AD device reached a PCE of 22.35%, a JSC of 24.57 mA cm−2, a VOC of 1.11 V, and an FF of 82.41%. The introduction of AD and ADA layers clearly improves the complete device performance of perovskite solar cells. With an appropriate ADA layer, the number of defects decreases, electron transfer is promoted more effectively, and charge recombination is suppressed.
| Sample | VOC (V) | JSC (mA cm−2) | FF (%) | PCE (%) |
|---|---|---|---|---|
| Pristine SnO2 | 1.06 | 24.52 | 78.16 | 20.22 |
| SnO2-AD | 1.11 | 24.57 | 82.41 | 22.35 |
| SnO2-ADA | 1.13 | 24.67 | 82.59 | 23.07 |
The PCE of the ADA-modified device, 23.07%, is not the highest among all reported cage-like molecular interface modifications, but it is still higher than some similar small-molecule passivation strategies. The VOC, FF, PCE, and JSC of SnO2-ADA devices show high reproducibility, and the average values are higher than those of pristine SnO2 perovskite solar cells. Table 2 compares the photovoltaic performance of high-efficiency perovskite solar cells modified with cage-like molecular interfaces. The data indicate that ADA modification is a competitive approach for improving perovskite solar cells.
| Year | Device structure | VOC (V) | FF (%) | PCE (%) |
|---|---|---|---|---|
| 2025 | ITO/SnO2/ADA/PVK/HTL/Ag | 1.13 | 82.59 | 23.07 |
| 2023 | ITO/SnO2/DABCO/PVK/HTL/Ag | 1.17 | 81.60 | 23.60 |
| 2023 | ITO/NiOX/AAD/PVK/PC61BM/Ag | 1.13 | 83.72 | 22.21 |
| 2024 | FTO/TiO2/PVK/PHMG/HTL/Au | 1.16 | 83.53 | 25.42 |
| 2025 | ITO/SnO2/PVK/ADAI/HTL/Au | 1.19 | 83.79 | 25.13 |
| 2023 | ITO/SnO2/TMACl/PVK/HTL/Ag | 1.16 | 80.78 | 23.08 |
| 2025 | FTO/TiO2/SnO2/NbSe2/PVK/HTL/Ag | 1.15 | 82.11 | 24.05 |
| 2022 | ITO/SnO2/PVK/ATH/HTL/Au | 1.18 | 82.60 | 23.45 |
| 2023 | ITO/SnO2/HMTA/PVK/HTL/Ag | 1.13 | 80.00 | 22.37 |
The EQE of SnO2-ADA devices is higher than that of SnO2-AD and pristine SnO2 devices. The enhancement of EQE means an efficient and fast photon-to-current conversion process, which is reflected in a slight increase in JSC. The comparison shows that both ADA and AD can suppress non-radiative recombination and improve charge extraction velocity. The integrated current density from EQE is consistent with the J–V results, confirming the reliability of the photovoltaic measurements for perovskite solar cells.
The water contact angle of the ADA-modified film increased from 50.85° to 59.83°, indicating that the quality of the perovskite film optimized by ADA is improved and its hydrophobicity is enhanced. After 1000 h of testing, the device still maintained more than 85% of its initial PCE. This demonstrates that the introduction of ADA significantly improves the long-term stability of perovskite solar cells. The enhanced stability is attributed to the passivation of interfacial defects, the reduced number of ionic migration pathways, and the hydrophobic nature of the adamantane cage, which limits moisture ingress.
The morphology of the ETL, PVK, and perovskite solar cells was investigated. The roughness of SnO2-ADA was 3.52 nm, which is almost unchanged compared with pristine SnO2 (3.72 nm), indicating that the ADA layer does not alter the overall morphology of the SnO2 layer. The PVK layer deposited on pristine SnO2 had an irregular morphology with many pinhole defects. In contrast, the PVK layer deposited on SnO2-ADA showed dense and flat grain growth with fewer pinholes and other defects. The cross-sectional morphology of the ITO/SnO2-ADA/PVK/HTL/Ag device showed no obvious change in film thickness but was relatively uniform. These visual morphology results indicate that coating ADA on the SnO2 surface can enhance adhesion between SnO2 and the PVK layer, enabling dense stacking and growth of perovskite polycrystals at the interface.
The crystallinity and fluorescence spectra of PVK films were also studied. The diffraction peak positions of the two samples are basically the same. However, the PbI2 peak intensity of the ADA-modified film decreased significantly, indicating that ADA passivates PbI2. Compared with the control PVK layer, the PVK crystalline peak intensity after AD and ADA modification became stronger, indicating improved PVK layer quality. The PL intensity of the PVK layer on SnO2-ADA was lower than that on pristine SnO2. The decrease in PL intensity is attributed to a quenching effect, indicating that the PVK layer on SnO2-ADA has higher electron extraction ability.
The TRPL parameters of perovskite films are listed in Table 3. Here, τ1 is the fast decay component, τ2 is the slow decay component, A1 is the weight factor of the fast decay component, A2 is the weight factor of the slow decay component, and τave is the average carrier lifetime. The average carrier lifetime is calculated by
$$ \tau_{\text{ave}} = \frac{A_1 \tau_1^2 + A_2 \tau_2^2}{A_1 \tau_1 + A_2 \tau_2} $$
The PVK layer on SnO2-ADA showed a shortened carrier lifetime of τave = 68.12 ns, indicating that the PVK layer can effectively transfer electrons to the SnO2-ADA electron transport layer. This proves that introducing adamantane-containing molecules into the PVK layer on pristine SnO2 helps reduce carrier lifetime, facilitates charge separation, and effectively suppresses recombination, resulting in higher PCE and VOC in perovskite solar cells. From SnO2-AD to SnO2-ADA, the defect passivation effect based on cage-like molecules gradually increases, which is positively correlated with the chemical interaction strength.
| Sample | τ1 (ns) | τ2 (ns) | A1 (%) | A2 (%) | τave (ns) |
|---|---|---|---|---|---|
| PVK on pristine SnO2 | 19.25 | 114.29 | 53.85 | 46.15 | 98.60 |
| PVK on SnO2-AD | 15.08 | 102.06 | 60.80 | 39.20 | 85.85 |
| PVK on SnO2-ADA | 8.42 | 85.45 | 74.67 | 25.33 | 68.12 |
The passivation mechanism and energy level alignment were analyzed by XPS and UPS. After ADA modification, the O 1s peak shifted slightly to higher binding energy, from 530.4 to 530.7 eV, indicating a strong interaction between SnO2 and ADA. The SnO2-ADA sample showed a high-intensity O 1s peak at 530.7 eV, which is assigned to lattice oxygen in SnO2, and another peak at 531.7 eV from oxygen vacancies. The lattice oxygen intensity of SnO2-ADA was slightly higher than that of pristine SnO2, while the oxygen vacancy defect signal was lower than that of pristine SnO2. This indicates that oxygen vacancies on the SnO2 surface can be passivated by Cl through electrostatic coupling. The Cl 2p peaks at 197.4 eV and 199.0 eV are assigned to Cl 2p3/2 and Cl 2p5/2, respectively, while no Cl 2p peak was found in pristine SnO2.
The Sn 3d5/2 and Sn 3d3/2 peaks of SnO2-ADA shifted in binding energy, indicating a strong interaction between ADA and SnO2. For Pb 4f, the pristine perovskite layer showed characteristic peaks at 137.93 eV and 142.78 eV. After introducing the ADA layer, the Pb 4f peaks shifted to higher binding energies of 138.21 eV and 143.22 eV, confirming a strong interaction between ADA and Pb2+. The PbI2 peak intensity of the ADA-modified film decreased markedly, further indicating that ADA passivates PbI2. In addition, the electron cloud density of the ADA molecule is mainly distributed on the amine group, indicating that the Lewis basic amine group can coordinate with under-coordinated Pb2+ in the PVK, reducing the Pb2+ vacancy density, suppressing residual PbI2, and improving perovskite crystallization quality. This demonstrates the defect passivation effect of ADA between the ETL and PVK layers, which is beneficial for reducing trap concentration and increasing carrier lifetime in perovskite solar cells.
The optical transmittance of the two samples is basically the same, indicating that the introduction of ADA does not affect the overall transmittance. The energy level analysis shows that when no ADA layer is added, electrons extracted from the PVK layer and transferred to SnO2 easily flow back to the perovskite layer and recombine with holes. When the ADA layer is introduced, because the ADA layer is very thin, electrons in the perovskite conduction band tunnel to the ETL, while ADA acts as a barrier layer that hinders reverse transport of electrons from the ETL. This blocks carrier recombination and increases the VOC of perovskite solar cells. The electron reflux barrier concept is thus realized by an ultrathin insulating-like cage molecular layer that still permits forward tunneling.
The electrochemical performance was further tested. In the Nyquist plot, compared with pristine SnO2, the SnO2-ADA device showed a lower transfer resistance and a higher recombination resistance, indicating that the ADA-modified device is more favorable for charge transfer and can effectively suppress non-radiative recombination. This is consistent with the increase in VOC. The Mott–Schottky analysis gives the built-in potential Vbi from
$$ C^{-2} = \frac{2}{A^2 \varepsilon \varepsilon_0 q N_D} (V_{bi} – V) $$
where C is capacitance, A is device area, ε is relative permittivity, ε0 is vacuum permittivity, q is elementary charge, ND is carrier density, and V is applied bias. After AD/ADA layer modification, the built-in potential increased from 0.92 V to 1.0 V. A higher Vbi can effectively prevent excessive charge accumulation. The VOC versus light intensity relationship reveals the recombination mechanism. The slope of VOC versus the natural logarithm of light intensity is given by
$$ V_{OC} = \frac{n k_B T}{q} \ln\left(\frac{J_{SC}}{J_0} + 1\right) $$
where n is the ideality factor, kB is the Boltzmann constant, T is temperature, JSC is short-circuit current density, and J0 is reverse saturation current density. Compared with pristine SnO2, the slope of the AD/ADA-modified device is smaller, indicating that trap-assisted non-radiative recombination is suppressed after AD/ADA modification. The theoretical maximum FF values of pristine SnO2 and ADA-modified devices were 82.3% and 85.3%, respectively, while the experimental values were only 78.2% and 82.6%. The ADA-modified device had significantly less charge transfer loss and non-radiative loss than the pristine SnO2 device, indicating that ADA modification effectively limits defect-induced non-radiative recombination and enhances charge transport in perovskite solar cells.
The transient photovoltage (TPV) curves show that SnO2-AD and SnO2-ADA devices have longer decay lifetimes than pristine SnO2, indicating that introducing ADA and AD reduces the internal defect density of the device. The combination of reduced defect density, improved charge extraction, and blocked electron reflux explains the simultaneous enhancement of VOC, JSC, FF, and PCE in perovskite solar cells.
To further quantify the photovoltaic performance, the PCE is calculated by
$$ \eta = \frac{V_{OC} J_{SC} FF}{P_{in}} \times 100\% $$
where Pin is the incident optical power density, typically 100 mW cm−2 under standard AM 1.5G illumination. Using the measured values for the SnO2-ADA device, VOC = 1.13 V, JSC = 24.67 mA cm−2, and FF = 82.59%, the calculated PCE is approximately 23.07%. This confirms the consistency of the photovoltaic parameters and the reliability of the efficiency measurement for perovskite solar cells.
The defect passivation and electron blocking mechanisms can be summarized as follows. First, the amino group in ADA coordinates with under-coordinated Pb2+ at the perovskite surface or grain boundaries, reducing Pb2+ vacancy defects and suppressing PbI2 residues. Second, the chloride ion in ADA interacts with oxygen vacancies on SnO2, reducing surface trap states and improving the ETL/perovskite interface quality. Third, the adamantane cage provides a rigid, hydrophobic, and thermally stable molecular framework that increases moisture resistance and long-term stability. Fourth, the ultrathin ADA layer acts as an electron reflux barrier: it allows forward electron tunneling from the perovskite to SnO2 but hinders reverse electron transport from SnO2 back to the perovskite, thereby reducing interfacial recombination and increasing VOC and FF in perovskite solar cells.
The optimal ADA concentration is a trade-off between passivation and charge transport. At 0.3 mg mL−1, the ADA coverage is sufficient to passivate defects and form a continuous ultrathin barrier layer without excessively increasing series resistance. At 0.5 mg mL−1, the ADA layer becomes too thick, suppressing charge transfer and reducing PCE. At 0.1 mg mL−1, the passivation effect is insufficient, and device performance decreases. Therefore, precise control of the interfacial modifier concentration is critical for high-performance perovskite solar cells.
The stability enhancement can be understood from both chemical and physical perspectives. Chemically, the strong coordination between ADA and Pb2+ and the electrostatic coupling between Cl− and oxygen vacancies stabilize the interface and reduce ion migration pathways. Physically, the hydrophobic adamantane cage increases the water contact angle and reduces moisture ingress into the perovskite film. The device retained more than 85% of its initial PCE after 1000 h in air, demonstrating that ADA modification improves the environmental stability of perovskite solar cells without sacrificing efficiency.
Table 4 summarizes the key interfacial and photovoltaic parameters for pristine, AD-modified, and ADA-modified devices. The data show a consistent trend: ADA modification gives the highest VOC, FF, and PCE, while maintaining a high JSC. The improvement is attributed to the combined effects of defect passivation, electron reflux blocking, and enhanced charge extraction.
| Parameter | Pristine SnO2 | SnO2-AD | SnO2-ADA |
|---|---|---|---|
| VOC (V) | 1.06 | 1.11 | 1.13 |
| JSC (mA cm−2) | 24.52 | 24.57 | 24.67 |
| FF (%) | 78.16 | 82.41 | 82.59 |
| PCE (%) | 20.22 | 22.35 | 23.07 |
| τave (ns) | 98.60 | 85.85 | 68.12 |
| Water contact angle (°) | 50.85 | — | 59.83 |
| PCE retention after 1000 h (%) | — | — | >85 |
The RMS roughness of SnO2-ADA was 3.52 nm, slightly lower than the 3.72 nm of pristine SnO2. Although the difference is small, the reduced roughness suggests better interfacial contact between the ETL and the perovskite layer. The SEM images showed that the perovskite film on SnO2-ADA had fewer pinholes and denser grains than that on pristine SnO2. The cross-sectional image showed a uniform film thickness and well-defined interfaces, which are beneficial for charge transport and device reproducibility in perovskite solar cells.
The XRD patterns of the perovskite films showed identical peak positions for the main perovskite phases, indicating that ADA modification does not change the crystal structure. However, the intensity of the PbI2 peak decreased after ADA modification, implying that residual PbI2 is reduced or passivated. The enhanced perovskite peak intensity suggests improved crystallinity and larger grain size, which are consistent with the reduced defect density and improved charge transport. The PL quenching and shortened TRPL lifetime further confirm that electron extraction from the perovskite to SnO2-ADA is more efficient than that to pristine SnO2.
The XPS results provide direct evidence for chemical interactions at the interface. The O 1s peak shift to higher binding energy indicates that the electron density around oxygen in SnO2 changes upon ADA adsorption, consistent with Cl− passivation of oxygen vacancies. The Sn 3d peak shift indicates a strong interaction between ADA and SnO2. The Pb 4f peak shift to higher binding energy confirms that the amine group in ADA coordinates with Pb2+, reducing the electron density around Pb and passivating under-coordinated Pb sites. These chemical interactions reduce interfacial trap states and suppress non-radiative recombination in perovskite solar cells.
The UPS analysis and energy level diagram show that the ADA layer introduces a thin tunneling barrier at the SnO2/perovskite interface. The conduction band offset is small enough for electrons to tunnel from the perovskite to SnO2, but the barrier is sufficient to block reverse electron flow from SnO2 to the perovskite. This asymmetric transport behavior is the origin of the electron reflux barrier effect. The built-in potential increases from 0.92 V to 1.0 V, which enhances the driving force for charge separation and reduces charge accumulation at the interface. As a result, the VOC and FF of perovskite solar cells are improved.
The EIS results show that the ADA-modified device has lower charge transfer resistance and higher recombination resistance than the pristine device. The lower charge transfer resistance indicates faster electron extraction from the perovskite to the ETL, while the higher recombination resistance indicates suppressed interfacial recombination. These impedance changes are consistent with the improved photovoltaic parameters. The TPV decay lifetime is longer for the ADA-modified device, further confirming reduced defect density and suppressed non-radiative recombination in perovskite solar cells.
The theoretical FF limit was calculated using the measured VOC and JSC values. For the pristine SnO2 device, the theoretical FF maximum was 82.3%, while the experimental FF was 78.2%. For the ADA-modified device, the theoretical FF maximum was 85.3%, while the experimental FF was 82.6%. The smaller gap between theoretical and experimental FF for the ADA-modified device indicates reduced charge transfer loss and non-radiative loss. This demonstrates that ADA modification effectively limits defect-induced non-radiative recombination and enhances charge transport in perovskite solar cells.
The long-term stability test showed that the ADA-modified device retained more than 85% of its initial PCE after 1000 h in air. The pristine device degraded faster, which is attributed to moisture-induced decomposition and ion migration at the buried interface. The hydrophobic adamantane cage and the strong interfacial coordination of ADA suppress moisture ingress and ion migration, thereby improving the stability of perovskite solar cells. The water contact angle increase from 50.85° to 59.83° provides direct evidence for the enhanced hydrophobicity of the ADA-modified film.
In summary, I introduced ADA as an interfacial modifier between SnO2 and the perovskite layer to form an electron reflux barrier in perovskite solar cells. The ADA molecule provides multiple functions: its amine group coordinates with under-coordinated Pb2+, its chloride ion passivates oxygen vacancies on SnO2, and its adamantane cage enhances hydrophobicity and structural stability. The ultrathin ADA layer permits forward electron tunneling but blocks reverse electron transport, reducing interfacial recombination and increasing VOC and FF. The optimal ADA concentration of 0.3 mg mL−1 yielded a PCE of 23.07%, a VOC of 1.13 V, a JSC of 24.67 mA cm−2, and an FF of 82.59%. The modified devices retained more than 85% of their initial PCE after 1000 h in air. These results demonstrate that cage-like molecular interface engineering is an effective strategy for improving both the efficiency and stability of perovskite solar cells.
Future work on perovskite solar cells can further optimize the ADA concentration, explore other adamantane derivatives with different functional groups, and combine ADA modification with other passivation strategies. The electron reflux barrier concept may also be extended to other device architectures and wide-bandgap perovskites. The combination of defect passivation, energy level alignment, and interfacial blocking in a single ultrathin molecular layer provides a promising route for high-performance and stable perovskite solar cells.
The key formulas and relationships used in this work are summarized below for clarity. The PCE is given by
$$ \eta = \frac{V_{OC} J_{SC} FF}{P_{in}} \times 100\% $$
The average carrier lifetime is
$$ \tau_{\text{ave}} = \frac{A_1 \tau_1^2 + A_2 \tau_2^2}{A_1 \tau_1 + A_2 \tau_2} $$
The Mott–Schottky relation is
$$ C^{-2} = \frac{2}{A^2 \varepsilon \varepsilon_0 q N_D} (V_{bi} – V) $$
The light-intensity dependence of VOC is
$$ V_{OC} = \frac{n k_B T}{q} \ln\left(\frac{J_{SC}}{J_0} + 1\right) $$
The fill factor is defined as
$$ FF = \frac{V_{MPP} J_{MPP}}{V_{OC} J_{SC}} $$
where VMPP and JMPP are the voltage and current density at the maximum power point. These equations provide a quantitative framework for evaluating the photovoltaic performance of perovskite solar cells and for understanding the improvements achieved by ADA interfacial modification.
The experimental results consistently show that the ADA-modified perovskite solar cells outperform both pristine and AD-modified devices. The enhancements in VOC, JSC, FF, and PCE are supported by morphology, crystallinity, PL, TRPL, XPS, UPS, EIS, and Mott–Schottky analyses. The electron reflux barrier formed by the ultrathin ADA layer is a key factor in reducing interfacial recombination and improving charge collection. The hydrophobic adamantane cage also contributes to long-term stability. Therefore, ADA interfacial modification is a promising approach for the development of efficient and stable perovskite solar cells.