Interface engineering via pre-engineered black phosphorus quantum dots for highly efficient carbon-based hole-transport-layer-free perovskite solar cells

Abstract

Planar, carbon-electrode-based perovskite solar cells (C-PSCs) without a hole transport layer (HTL) are highly attractive due to their simple fabrication, low cost, and scalability. However, their performance is often limited by inefficient physical and electrical contact at the perovskite/carbon interface, which impedes hole extraction and promotes charge recombination. This study introduces a pre-engineered, multifunctional interlayer for HTL-free C-PSCs utilizing tetrabutylammonium ion (TBA⁺)-intercalated black phosphorus quantum dots (BPQDs). The TBA⁺ intercalation during synthesis pre-engineers the BPQDs with enhanced conductivity, a raised valence band maximum (−5.27 eV), and defect-passivation capabilities. This creates a favorable cascade energy-level alignment between the perovskite absorber (−5.5 eV) and the carbon electrode (−5.0 eV), thereby facilitating efficient hole extraction. The BPQDs interlayer also ensures seamless perovskite/carbon contact, promoting interfacial charge transfer. Additionally, TBA⁺ ions released from BPQDs effectively passivate defects on the perovskite surface, suppressing nonradiative recombination. Consequently, the optimized devices achieve a power conversion efficiency (PCE) of 17.08%, which is 24.1% and 11.9% higher than that of control devices without an interlayer (13.76%) and with a pristine BPQDs interlayer (15.26%), respectively. Furthermore, the encapsulated devices demonstrate improved operational stability, retaining 89.1% of their initial PCE after 360 hours under 1-sun illumination at 85 °C and 85% relative humidity.

---

New concepts

This work introduces a transformative materials design paradigm and an active interface-engineering concept that moves beyond conventional interface modification strategies for planar HTL-free C-PSCs. The electrochemical intercalation-assisted liquid-phase exfoliation (EIA-LE) method is presented not just as a faster synthesis route, but as a holistic protocol that intrinsically builds functionality into the 2D material. Specifically, during the production of black phosphorus quantum dots (BPQDs), we intercalate tailored tetrabutylammonium (TBA⁺) cations into their lamellar layers. This “synthesis-to-integration” concept means that the BPQDs are born ready-to-use, with their final desired properties (optimal energy levels and conductivity) pre-engineered. When used as an interface modifier, the “ion-anchored” BPQDs interlayer acts not merely as a physical connector but as active, electronically coupled bridges, resulting in seamless perovskite/BPQDs/carbon contacts that are simultaneously optimal for mechanical adhesion, electrical conduction, energy level alignment, and surface defect passivation (by released TBA⁺ ions from BPQDs). Differentiating from the paradigm of seeking pure materials, this “functional impurity” concept transforms the synthesis process itself into a direct engineering tool for creating tailored materials with built-in, multi-role capabilities, offering a generalizable pathway for streamlined interface design in optoelectronics.

Introduction

Perovskite solar cells (PSCs) have witnessed remarkable progress over the past decade, with their power conversion efficiencies (PCEs) skyrocketing from 3.8% to over 27%.^1–3 This rapid advancement has established PSCs as a frontrunner in photovoltaic technology. Nevertheless, their transition from laboratory-scale innovation to large-scale practical application faces significant barriers, primarily limited by insufficient long-term operational stability and the reliance on expensive organic hole-transport materials (HTMs, e.g., Spiro-OMeTAD) and precious metal electrodes (e.g., gold).^4,5

Carbon-electrode-based perovskite solar cells (C-PSCs) have emerged as a promising solution to these challenges, providing enhanced stability, lower cost, and excellent scalability.^6–8 The p-type nature of the carbon electrode further allows for a simplified, hole transport layer (HTL)-free architecture by eliminating the need for unstable and expensive organic HTL, which has generated significant research interest.^9,10 However, while mesoscopic HTL-free C-PSCs have recently achieved PCE over 23%,¹¹ the performance of planar equivalents remains inferior (Table S1), largely owing to poor physical and electrical contact at the perovskite/carbon interface.^12,13 Inserting an inorganic interlayer, such as perovskite quantum dots (QDs), metal chalcogenides (e.g., CdSe, PbS), carbon QDs, or two-dimensional (2D) materials (e.g. MoS₂, MXenes, black phosphorus), has recently emerged as a promising strategy to boost the PCE of planar HTL-free C-PSCs while retaining their inherent advantages.^14–21

Among the 2D material family, black phosphorus (BP) has gained significant attention due to its ambipolar charge transport, high carrier mobility, and tunable bandgap.^22–24 Notably, BP exhibits a layer-dependent bandgap that can be readily tuned from 2.0 eV in bulk form to 0.3 eV in monolayer phosphorene.²⁵ The room-temperature carrier mobility of bulk BP is as high as 350 cm² V⁻¹ s⁻¹ for holes and 220 cm² V⁻¹ s⁻¹ for electrons, while few-layer BP nanosheets (up to ∼10 nm thick) can exhibit mobilities of up to 1000 cm² V⁻¹ s⁻¹.^24,26 Furthermore, the bandgap and carrier mobility of BP can be further modulated through strain engineering, functionalization, doping, etc.^27–29 Owing to these intriguing and tunable semiconducting properties, 2D BP can serve multiple functions in photovoltaics, including regulating carrier dynamics, tailoring absorber crystallization, improving energy-level alignment, and enhancing light harvesting.^29–34 For instance, Zhang et al. modified both the electron transport layer/perovskite and perovskite/HTL interfaces with thickness-tailored BP nanosheets, achieving synergistic cascade carrier extraction and boosting the PCE from 16.95% to 19.83% in metal-electrode-based PSCs.³³ Myagmarsereejid et al. established favorable band energy alignment at the perovskite/carbon interface using large-area BP flakes, enabling efficient hole extraction and a PCE of 15.58% for C-PSCs.³⁴ However, the sparse distribution of 2D BP nanosheets commonly deposited by spin-coating and drop-casting fails to fully exploit the ability of BP as an interface modifier. In contrast to BP nanosheets and flakes, BP quantum dots (BPQDs) offer greater potential for modifying the rough and loose perovskite/carbon contact in planar HTL-free C-PSCs. Moreover, the laminated structure of BP provides a promising platform for functional expansion by pre-engineering the materials during synthesis. For photovoltaic applications, embedding functional ions or molecules within the interlamellar spaces of BP could endow the film with effective passivation capabilities, in addition to improving band alignment and charge extraction. Nevertheless, this “synthesis-to-integration” strategy remains largely unexplored for PSCs.

In this work, we developed a facile synthesis of pre-engineered, multifunctional black phosphorus quantum dots (BPQDs) via an electrochemical-intercalation-assisted liquid-phase exfoliation (EIA-LE) method in a tetrabutylammonium chloride (TBACl) electrolyte. Integrated as an interlayer between the perovskite and carbon electrode, these BPQDs provide seamless electrical contact and optimal energy level alignment, which facilitates highly efficient hole extraction. Furthermore, the TBA⁺ cations retained within the BPQDs play a dual role: they enhance the interlayer's conductivity for rapid hole transport and passivate perovskite surface defects to minimize non-radiative recombination. As a result, the optimized planar HTL-free C-PSCs deliver a high open-circuit voltage (V_oc) of 1.05 V and a boosted champion PCE of 17.08%, up from 13.76% for the control devices. The encapsulated devices also showed robust operational stability, maintaining 89.1% of their initial PCE after 360 hours of aging under 1-sun illumination at 85 °C and 85% relative humidity.

Results and discussion

Currently, BP nanosheets and QDs are typically prepared via liquid-phase ultrasonic exfoliation of bulk BP crystals. However, this process requires several to tens of hours, increasing production cost and often introducing structural defects in the final product.^29,35Fig. 1a schematically illustrates the process for BPQDs synthesis via the EIA-LE approach. A typical three-electrode system with a TBACl (0.1 M)/propylene carbonate (PC) electrolyte is used for the electrochemical intercalation process, where a bulk BP crystal foil, platinum foil, and a saturated calomel electrode serve as the working, counter, and reference electrodes, respectively (Fig. S1a). Applying a potential of −8.0 V drives TBA⁺ cations toward the cathode. Concurrent cathodic reduction leads to the formation of H₂ (from the decomposition of PC and/or residual H₂O) at the BP interface. The evolution of H₂ bubbles disrupts van der Waals forces, enlarges the laminate gaps of BP, and thereby facilitates the intercalation of TBA⁺ into the BP layers,^36,37 as evidenced by the significant volume expansion of the BP foil (Fig. S1). The expanded BP was carefully removed from the electrolyte, rinsed with isopropanol (IPA), and then subjected to mild ultrasonic exfoliation (400 W) in IPA for 30 min. Notably, the EIA-LE synthesis requires only ∼1 h. For comparison, BPQDs were also prepared via traditional liquid-phase ultrasonic exfoliation (600 W at 0 °C for 10 h, see Experimental section). The BPQDs dispersions were finally collected by centrifuging the mixtures at 6000 rpm for 15 min to remove large aggregates and nanosheets. For short, the BPQDs prepared by EIA-LE and direct ultrasonic exfoliation are denoted as BP-TBACl and BP, respectively.

Fig. 1 (a) Schematic illustration of the BP-TBACl preparation process via electrochemical-intercalation-assisted liquid-phase exfoliation (EIA-LE). (b) and (c) TEM and HRTEM images of (b) pristine BP and (c) BP-TBACl. (d) Corresponding size distribution histograms and (e) Raman spectra for both samples.

The microstructures of the two BPQDs were observed by transmission electron microscopy (TEM) and high-resolution TEM (HRTEM). TEM images shown in Fig. 1b and c confirm the successful preparation of PBQDs via both methods, with a similar average diameter of ∼2.5 nm (Fig. 1d). The HRTEM image of the control BP (inset of Fig. 1b) shows a lattice spacing of 0.215 nm, corresponding to the (002) crystal planes of orthorhombic phosphorus (ICDD-PDF: no. 76-1963).³² In contrast, well-defined lattice fringes are difficult to observe for BP-TBACl (inset of Fig. 1c), likely due to the random insertion of TBA⁺ ions, which generates steric stress and induces local strain,³⁸ distorting the original lattice structure. Raman spectroscopy of the control BP shows three prominent peaks at 350, 410, and 467 cm⁻¹ (Fig. 1e), which can be assigned to the out-of-plane phonon mode A¹_g and two in-plane modes B²_g and A²_g, respectively.³⁹ Compared to the control, BP-TBACl exhibits a slight red shift in all three Raman modes, which can be attributed to ion insertion and absorption restricting the vibrational freedom of phosphorus atoms, thereby reducing the energy of Raman scattering.⁴⁰

X-ray photoelectron spectroscopy (XPS) was performed to further investigate the interaction between TBACl and BP. As shown in Fig. 2a, the XPS peak centered at a binding energy of 134.0 eV is associated with P 2p, which is the key feature of highly crystalline BP.⁴¹ A distinct N 1s peak at 402.0 eV is observed for BP-TBACl (Fig. 2b), confirming the successful intercalation of TBA⁺ ions.⁴² In contrast, only a negligible N 1s signal is detected for pristine BP, which can be attributed to the absorbed N-methylpyrrolidone (NMP, the solvent used in the traditional exfoliation method). Density functional theory calculations reveal that Cl⁻ has a much lower adsorption energy (−0.83 eV) on BPQDs than TBA⁺ (−1.78 eV) (Fig. 2c and Table S2), suggesting poor adsorption. This is consistent with the weak Cl 2p signal in the XPS spectrum of BP-TBACl (Fig. S2). This difference is primarily due to the substantial van der Waals interactions between the organic framework of the TBA⁺ and the BP surface.⁴³

Fig. 2 High-resolution XPS spectra of (a) P 2p and (b) N 1s for pristine BP and EIA-LE processed BP-TBACl. (c) Adsorption configurations and corresponding adsorption energies for TBA⁺ and Cl⁻ on BP. (d) Theoretically calculated energy band structure and projected density of states (PDOS) for BP-TBACl. (e) UV-vis absorption spectra of BP and BP-TBACl in isopropanol, with an inset showing a digital photograph of a stable BP-TBACl dispersion. (f) I–V curves of BP and BP-TBACl films; the inset illustrates the device architecture used for testing.

Theoretical calculations of the electronic band structure and projected density of states (PDOS) shown in Fig. 2d indicate a bandgap of approximately 1.70 eV for BP-TBACl,³² consistent with the value of 1.62 eV obtained from ultraviolet-visible absorption measurements (Fig. 2e and Fig. S3). PDOS analysis of different ions adsorbed on BP (Fig. 2d, right panel) indicates that the strong p-type doping effect induced by TBA⁺ is the dominant factor altering the electronic structure of the pre-engineered BP-TBACl, particularly the valence band.^44,45 The influence of Cl⁻ is negligible. Fig. 2f displays the representative I–V curves of Ag/BP (or BP-TBACl)/Ag devices. The BP-TBACl film exhibits higher conductivity, demonstrating its improved carrier transport capability, which is crucial for photovoltaic applications.⁴⁶

The morphologies of pristine MAPbI₃ (denoted as MA) and BP-TBACl decorated MAPbI₃ (MA-BP-TABCl) films were characterized by scanning electron microscopy (SEM). As depicted in Fig. 3a, the pristine MA film exhibits clear and well-defined grains. After modification, a homogeneous layer of BPQDs covers the perovskite surface (Fig. 3b). Atomic force microscopy (AFM) further examined the surface topography. The perovskite surface becomes smoother after spin-coating with BP-TBACl (Fig. 3c and d). The root mean square (RMS) roughness decreases from 13.2 nm to 9.6 nm after BPQDs modification (Fig. 3e and f). Cross-sectional SEM images (Fig. 3g) reveal voids and cracks at the pristine perovskite/carbon interface, which would severely impede interfacial charge transfer.^47,48 In contrast, the thin BP-TBACl interlayer effectively bridges the perovskite and carbon layers, establishing an intimate interfacial contact (Fig. 3h). I–V measurements of devices with an ITO/MA-BP-TBACl (or MA)/carbon structure (Fig. S5) show higher conductivity for the MA-BP-TBACl sample due to the formation of seamless perovskite/BPQDs/carbon contacts. These improved interfacial contacts are conducive to enhancing the hole transfer.⁴⁹ UV-vis absorption spectra (Fig. S6a) show no significant changes among the MA, MA-BP, and MA-BP-TBACl films, indicating that the BPQDs modification does not affect the optical properties of the perovskite layer.

Fig. 3 (a) and (b) Top-view SEM, (c) and (d) AFM, and (e) and (f) corresponding 3D-AFM images of pristine MA and MA-BP-TBACl films, respectively. Cross-sectional SEM images of the complete HTL-free C-PSCs (g) without and (h) with a BP-TBACl interlayer.

To gain insights into the effects of the two BPQDs on the perovskite surface state and charge transport dynamics, we examined the optoelectronic properties of various electrodes (MA/carbon, MA-BP/carbon, and MA-BP-TBACl/carbon on quartz) using steady-state photoluminescence (PL) and time-resolved PL (TRPL) spectrometry. As depicted in Fig. 4a, the steady-state PL intensity is significantly quenched for the BPQDs/carbon films, especially for pre-engineered MA-BP-TBACl, indicating markedly enhanced hole extraction from the perovskite to the carbon electrode when the BP-TBACl interlayer is present. The TRPL decays (Fig. 4b) were fitted with a bi-exponential function, containing a fast decay component (associated with bimolecular quenching/radiative recombination) and a slow decay component (associated with trapping-induced monomolecular/non-radiative recombination).^32,50 The obtained dynamic parameters (Table S3) reveal that the fast decay lifetime (τ₁) and slow decay lifetime (τ₂) for the MA/carbon device are 8.1 and 542.1 ns, respectively, corresponding to an average lifetime (τ_avg) of 531.6 ns. In contrast, MA-BP-TBACl exhibits a significant decrease in τ₁ and τ₂ (5.3 and 370.2 ns, respectively), with a τ_avg (364.1 ns) much lower than that of MA/carbon, indicating faster transfer of photogenerated holes from the absorber to the conductive carbon electrode due to the high hole mobility, favorable cascade band alignment, and intimate contacts facilitated by the BP-TBACl interlayer.

Fig. 4 (a) Steady-state PL spectra and (b) time-resolved PL decays of MA, MA-BP, and MA-BP-TBACl films coated with carbon electrodes. (c) Space-charge-limited-current (SCLC) measurements of hole-only devices. (d) Steady-state PL spectra of MA, MA-BP, and MA-BP-TBACl films. (e) Schematic illustrating the defect-passivation mechanism of BP-TBACl on perovskite. (f) Mott–Schottky plots of the corresponding complete C-PSCs.

Space charge limited current (SCLC) measurements were conducted on hole-only devices (ITO/NiO₂/perovskite/carbon) with and without an interlayer to estimate the charge trap densities (N_t).⁵¹ As shown in Fig. 4c, the MA-BP-TBACl-based device has the lowest trap-filled limit voltage (V_TFL) of 0.88 V, compared to 0.91 V for MA-BP and 0.99 V for the MA-based ones. The calculated N_t values are 2.72 × 10¹⁶, 2.50 × 10¹⁶, and 2.42 × 10¹⁶ cm⁻³ for MA, MA-BP, and MA-BP-TBACl films, respectively. This confirms the excellent defect passivation ability of the BPQDs, consistent with previous reports.^29,31,33 Furthermore, the TBA⁺ cations released from pre-engineered BP-TBACl can also passivate A-site vacancies by forming 1D perovskite (i.e., TBAPbI₃), while Cl⁻ anions can fill I vacancies,^50,52–54 further reducing the N_t of MA-BP-TBACl films. The increased PL intensity of the perovskite films after BPQDs modification also confirms their passivation capacity (Fig. 4d). A schematic of the proposed defect passivation mechanism is shown in Fig. 4e. The lower N_t is crucial for suppressing non-radiative recombination and improving device performance.⁵⁵ Mott–Schottky analysis was employed to estimate the built-in potential (V_bi) of the C-PSCs. As shown in Fig. 4f, the V_bi for the MA-BP-TBACl-based device (0.42 V) is higher than that of the MA-BP (0.38 V) and the MA-based (0.35 V) ones. A higher V_bi indicates a stronger driving force for charge carrier separation and transport within the solar cells.⁵⁶

The energy levels of MAPbI₃, BP, and BP-TBACl films were determined by UV-vis spectroscopy and ultraviolet photoelectron spectroscopy (UPS). The valence band maximum (VBM) values are tested to be −5.50, −5.33, and −5.27 eV for MAPbI₃, BP, and BP-TBACl, respectively (Fig. S7).⁵⁷ The band gaps (E_g) determined from Tauc plots (Fig. S3 and S6b) are 1.55, 1.72, and 1.62 eV for MAPbI₃, BP, and BP-TBACl, respectively. Therefore, the conduction band minimum (CBM) is calculated to be −3.95, −3.61, and −3.65 eV for MA, BP, and BP-TBACl, respectively. In this study, planar HTL-free C-PSCs with an architecture of ITO/SnO₂/MAPbI₃/BPQDs (or without)/carbon (Fig. 5a) were fabricated. The energy level alignments depicted in Fig. 5b show that both BPQDs interlayers create a cascade alignment between the perovskite and carbon electrode. Notably, for the pre-engineered BP-TBACl-based device, the VBM offsets at both the MAPbI₃/BP-TBACl and BP-TBACl/carbon interfaces are less than 0.3 eV, which is highly beneficial for accelerating hole extraction, consistent with the PL and TRPL results (Fig. 4a and b). This favorable band alignment reduces the thermodynamic voltage loss of the final solar cells.^58–60

Fig. 5 (a) Schematic of the HTL-free C-PSCs structure with a BP-TBACl interlayer. (b) Corresponding band alignment, (c) J–V curves, (d) EQE spectra, and (e) statistical distribution of PCE for devices based on MA, MA-BP, and MA-BP-TBACl films. (f) Stability of unencapsulated C-PSCs with and without MA-BP-TBACl stored in a N₂ atmosphere. (g) Operational stability of encapsulated devices under continuous 1-sun illumination at 85 °C/85% relative humidity.

The current density–voltage (J–V) curves of the champion planar HTL-free C-PSCs are presented in Fig. 5c and Fig. S8, with detailed photovoltaic parameters summarized in Table S4. The control MA-based C-PSC yields a PCE of 13.76%, with a short-circuit current density (J_sc) of 21.94 mA cm⁻², an open-circuit voltage (V_oc) of 0.99 V, and a fill factor (FF) of 0.634. Incorporating a directly ultrasonically exfoliated BP interlayer increases the PCE to 15.26%. Impressively, the MA-BP-TBACl-based devices achieve a champion PCE of 17.08% (J_sc = 23.57 mA cm⁻², V_oc = 1.05 V, FF = 0.689), underscoring the superiority of this EIA-LE-processed, pre-engineered interlayer. The PCE and V_oc values are competitive compared to reported planar HTL-free C-PSCs employing MAPbI₃ absorbers (Table S1). The performance enhancement stems from multiple factors. The optimal cascade band alignment (Fig. 5b) and reduced trap density (Fig. 4c–e) promote efficient hole extraction and suppress non-radiative recombination, boosting the V_oc and FF. Concurrently, the tight perovskite/BP-TBACl/carbon interface, with its improved conductivity (Fig. S5), facilitates charge transport, further increasing the FF. These combined improvements in carrier extraction, transfer, and transport also lead to improved charge collection efficiency and thereby the significant rise in J_sc. The integrated J_sc values from external quantum efficiency (EQE) spectra (Fig. 5d) are 21.01, 22.36, and 23.14 mA cm⁻² for MA, MA-BP, and MA-BP-TBACl-based devices, respectively, which are consistent with the results of J–V measurements. Photovoltaic performance under reverse and forward scanning reveals a significantly reduced hysteresis index of 0.174 for the MA-BP-TBACl device, compared to 0.242 and 0.438 for the MA-BP- and MA-based ones (Fig. S8 and Table S4). This further confirms accelerated charge transfer and reduced interfacial charge accumulation in the devices with BPQDs interlayers, especially for the MA-BP-TBACl-based one. The statistical distribution of PCE values (Fig. 5e) demonstrates the excellent reproducibility of the MA-BP-TBACl-based devices.

To further demonstrate the advantage of the pre-engineered BP-TBACl interlayer, large-area, HTL-free C-PSCs were fabricated. The corresponding J–V characteristics are presented in Fig. S9 and Table S5. The 1 cm² solar cells with the BP-TBACl interlayer achieve a PCE of 13.52%, with a J_sc of 22.78 mA cm⁻², a V_oc of 1.05 V, and a FF of 0.565. The decrease in PCE compared to the 0.04 cm² devices is primarily due to the reduced FF, which is dominated by the higher series resistance of the large-area carbon electrode. For comparison, the PCE degradation is more pronounced in 1 cm² cells without an interlayer (7.84%) and with a traditional BP interlayer (11.45%). These results confirm the superior capability of the pre-engineered BP-TBACl interlayer in improving interfacial hole extraction and transfer.

The long-term durability of the obtained C-PSCs was evaluated. As shown in Fig. 5f and Fig. S10, the unencapsulated MA-BP-TBACl device retains 86.5% of its initial PCE after 700 h of storage in an N₂ atmosphere, a result of effective defect passivation and improved interfacial contact. In contrast, the control device without an interlayer suffers a significant efficiency loss of 21.8%. Furthermore, under continuous 1-sun illumination and damp-heat conditions (85 °C and 85% relative humidity), the encapsulated MA-BP-TBACl-based C-PSCs exhibit enhanced operational stability, retaining 89.1% of their initial PCE after 360 h (Fig. 5g and Fig. S11). Under the same conditions, the MA-based device lost 18.2% of its initial performance. This finding indicates that the BP-TBACl interlayer also helps to mitigate thermal stress, thereby enhancing the photothermal stability of the encapsulated devices.

Conclusions

In summary, we have developed a “synthesis-to-integration” strategy to produce high-quality, pre-engineered BPQDs with multifunctional properties via a facile electrochemical-intercalation-assisted liquid-phase exfoliation technique. The intercalation of TBA⁺ cations enables fine BPQDs with enhanced conductivity and a raised valence band maximum. When integrated as an interlayer in planar HTL-free C-PSCs, these BPQDs establish a seamless interfacial contact and a favorable cascade energy alignment between the perovskite and carbon electrode, thereby facilitating efficient charge extraction and transfer. Concurrently, the TBA⁺ and Cl⁻ ions from TBACl effectively passivate surface defects, suppressing nonradiative recombination. The optimized device achieves a high V_oc of 1.05 V and a PCE of 17.08%, coupled with excellent stability, retaining over 89.1% of initial efficiency after 360 hours in damp-heat conditions. This work demonstrates that pre-engineering 2D materials with functional ions ab initio is a powerful and generalizable approach. It opens a pathway for tailoring materials to specific interfaces, effectively streamlining the journey from material discovery to high-performance optoelectronic devices.

Author contributions

Yipng Zhang and Xinwei Li: conceptualization, investigation, data collection and curation, writing – original draft. Aohan Mei: preparation of BPQDs. Guoge Zhang: methodology, validation. Shenghuang Lin: Resources, validation, supervision, writing – review & editing. Jun Du: supervision, validation. Nianqing Fu: resources, funding acquisition, supervision, validation, writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

All the relevant data are available from the corresponding authors upon request.

Supplementary information (SI) including experimental details; photos of the electrochemical intercalation setup; additional XPS spectra, Tauc plots, and UPS spectra of various BPQDs; conductivity curves and UV-vis spectra of various BPQDs/perovskite films; additional I–V curves, photovoltaic parameters, and stability results of PSCs. See DOI: https://doi.org/10.1039/d5mh01839k.

Acknowledgements

This work was financially supported by the Guangdong Basic and Applied Basic Research Foundation (Grants No. 2023A1515030006), National Natural Science Foundation of China (Grants No. 62374059).

References

1. J. Du, J. Chen, B. Ouyang, A. Sun, C. Tian, R. Zhuang, C. Chen, S. Liu, Q. Chen, Z. Li, X. Wu, J. Cai, Y. Zhao, R. Li, T. Xue, T. Cen, K. Zhao and C.-C. Chen, Energy Environ. Sci., 2025, 18, 3196–3210.
2. S. Hu, J. Wang, P. Zhao, J. Pascual, J. Wang, F. Rombach, A. Dasgupta, W. Liu, M. A. Truong, H. Zhu, M. Kober-Czerny, J. N. Drysdale, J. A. Smith, Z. Yuan, G. J. W. Aalbers, N. R. M. Schipper, J. Yao, K. Nakano, S.-H. Turren-Cruz, A. Dallmann, M. G. Christoforo, J. M. Ball, D. P. McMeekin, K.-A. Zaininger, Z. Liu, N. K. Noel, K. Tajima, W. Chen, M. Ehara, R. A. J. Janssen, A. Wakamiya and H. J. Snaith, Nature, 2025, 639, 93–101.
3. X. Li, J. Wang, Y. Min, Q. Xue, J. Du and N. Fu, ACS Energy Lett., 2025, 10, 3203–3222.
4. X. Yin, Z. Song, Z. Li and W. Tang, Energy Environ. Sci., 2020, 13, 4057–4086.
5. N. Li, X. Niu, Q. Chen and H. Zhou, Chem. Soc. Rev., 2020, 49, 8235–8286.
6. Y. Lin, J. Tang, H. Yan, J. Lin, W. Wang, S. Zhang, H. Rao, Z. Pan and X. Zhong, Energy Environ. Sci., 2024, 17, 4692–4702.
7. W. Shao, Y. Fu, J. He, Z. Deng, Y. Li, L. Zhang, W. Wu, Y. Wu and W.-H. Zhu, Adv. Funct. Mater., 2025, e18907.
8. H. Chen and S. Yang, Adv. Mater., 2017, 29, 1603994.
9. J. He, Y. Bai, Z. Luo, R. Ran, W. Zhou, W. Wang and Z. Shao, Energy Environ. Sci., 2025, 18, 2136–2164.
10. J. Xiang, C. Han, Y. Cheng, Q. Gao, W. Hu, Y. Zhou, A. Mei, Y. Zhou and H. Han, Adv. Mater., 2025, 37, e2415405.
11. Y. Ma, J. Liu, X. Chen, X. Zhao, J. Qi, B. She, S. Liu, Y. Jiang, Y. Sheng, C. Han, G. Zhang, J. Xie, K. Chen, Y. Cheng, J. Xiang, L.-M. Yang, Y. Zhou, F. Ling, Y. Zhou, A. Mei and H. Han, Nat. Energy, 2025, 10, 1084–1094.
12. J. Zhang, N. Cheng, H. Zhou, J. Zhong, F. He, Y. Chen, C. Chen, Z. Liu and P.-A. Zong, Mater. Today Energy, 2024, 43, 101599.
13. J. Tang, Y. Lin, H. Yan, J. Lin, H. Rao, Z. Pan and X. Zhong, Angew. Chem., Int. Ed., 2024, 63, e202406167.
14. M. Yuan, X. Gu, Y. Wang, K. A. Khawaja, J. Cappola, C. Picart, L. Li and F. Yan, Adv. Funct. Mater., 2025, e15519.
15. K. Zou, Q. Li, J. Fan, H. Tang, L. Chen, S. Tao, T. Xu and W. Huang, ACS Mater. Lett., 2022, 4, 1101–1111.
16. W. Wang, J. Zhang, H. Guo, Z. Pan, H. Rao, G. Zhang and X. Zhong, ChemSusChem, 2024, 17, e202301761.
17. H. Tang, T. Xu, X. Qin, K. Zou, S. Lv, J. Fan, T. Huang, L. Chen and W. Huang, ACS Appl. Nano Mater., 2021, 4, 13339–13351.
18. S. S. Mali, J. V. Patil and C. K. Hong, Adv. Energy Mater., 2020, 10, 1902708.
19. W. Chen, K. Li, Y. Wang, X. Feng, Z. Liao, Q. Su, X. Lin and Z. He, J. Phys. Chem. Lett., 2017, 8, 591–598.
20. X. Shen, X. Lin, Y. Peng, Y. Zhang, F. Long, Q. Han, Y. Wang and L. Han, Nano-Micro lett., 2024, 16, 201.
21. A. S. R. Bati, M. Batmunkh and J. G. Shapter, Adv. Energy Mater., 2020, 10, 1902253.
22. M. Batmunkh, M. Bat-Erdene and J. G. Shapter, Adv. Energy Mater., 2018, 8, 1701832.
23. J. Pang, A. Bachmatiuk, Y. Yin, B. Trzebicka, L. Zhao, L. Fu, R. G. Mendes, T. Gemming, Z. Liu and M. H. Rummeli, Adv. Energy Mater., 2017, 8, 1702093.
24. M. Zhang, G. M. Biesold and Z. Lin, Chem. Soc. Rev., 2021, 50, 13346–13371.
25. V. Tran, R. Soklaski, Y. Liang and L. Yang, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 89, 235319.
26. L. Li, J. Kim, C. Jin, G. J. Ye, D. Y. Qiu, F. H. da Jornada, Z. Shi, L. Chen, Z. Zhang, F. Yang, K. Watanabe, T. Taniguchi, W. Ren, S. G. Louie, X. H. Chen, Y. Zhang and F. Wang, Nat. Nanotechnol., 2017, 12, 21–25.
27. J. Kim, S. S. Baik, S. H. Ryu, Y. Sohn, S. Park, B.-G. Park, J. Denlinger, Y. Yi, H. J. Choi and K. S. Kim, Science, 2015, 349, 4.
28. A. S. R. Bati, P. Myagmarsereejid, M. Fronzi, K. Fan, P. Liu, Y. L. Zhong, P. L. Burn, I. R. Gentle, P. E. Shaw and M. Batmunkh, Small Struct., 2024, 5, 2300334.
29. T. Liu, Y. Liu, M. Chen, X. Guo, S. Tang, R. Zhang, Z. Xie, J. Wang, A. Gu, S. Lin and N. Wang, Adv. Funct. Mater., 2022, 32, 2106779.
30. X. Gong, L. Guan, Q. Li, Y. Li, T. Zhang, H. Pan, Q. Sun, Y. Shen, C. Grätzel, S. M. Zakeeruddin, M. Grätzel and M. Wang, Sci. Adv., 2020, 6, eaay5661.
31. W. Chen, K. Li, Y. Wang, X. Feng, Z. Liao, Q. Su, X. Lin and Z. He, J. Phys. Chem. Lett., 2017, 8, 591–598.
32. N. Fu, C. Huang, P. Lin, M. Zhu, T. Li, M. Ye, S. Lin, G. Zhang, J. Du, C. Liu, B. Xu, D. Wang and S. Ke, J. Mater. Chem. A, 2018, 6, 8886–8894.
33. M. Zhang, M. Ye, W. Wang, C. Ma, S. Wang, Q. Liu, T. Lian, J. Huang and Z. Lin, Adv. Mater., 2020, 32, e2000999.
34. P. Myagmarsereejid, S. Suragtkhuu, Q. T. Trinh, T. Gould, N. T. Nguyen, M. Bat-Erdene, E. Campbell, M. T. Hoang, W.-H. Chiu, Q. Li, H. Wang, Y. L. Zhong and M. Batmunkh, npj 2D Mater. Appl., 2024, 8, 38.
35. M. Batmunkh, K. Vimalanathan, C. Wu, A. S. R. Bati, L. Yu, S. A. Tawfik, M. J. Ford, T. J. Macdonald, C. L. Raston, S. Priya, C. T. Gibson and J. G. Shapter, Small Methods, 2019, 3, 1800521.
36. N. Wang, N. Mao, Z. Wang, X. Yang, X. Zhou, H. Liu, S. Qiao, X. Lei, J. Wang and H. Xu, Adv. Mater., 2021, 33, 2005815.
37. A. Ambrosi and M. Pumera, Chem. Soc. Rev., 2018, 47, 7213–7224.
38. D. A. Prishchenko, V. G. Mazurenko, M. I. Katsnelson and A. N. Rudenko, 2D Mater., 2017, 4, 025064.
39. T. Liu, Y. Liu, M. Chen, X. Guo, S. Tang, R. Zhang, Z. Xie, J. Wang, A. Gu and S. Lin, Adv. Funct. Mater., 2022, 32, 2106779.
40. Y. Zhao, H. Wang, H. Huang, Q. Xiao, Y. Xu, Z. Guo, H. Xie, J. Shao, Z. Sun and W. Han, Angew. Chem., Int. Ed., 2016, 128, 5087–5091.
41. S.-I. Iida, M. Terashima, K. Mamiya, H.-Y. Chang, S. Sasaki, A. Ono, T. Kimoto and T. Miyayama, J. Vac. Sci. Technol., B, 2021, 39, 044001.
42. S. Ohisa, M. Suzuki, T. Chiba and J. Kido, ACS Appl. Mater. Interfaces, 2019, 11, 25351–25357.
43. F. Li, W. Chen, H. Yang, R. Hu, Z. Huang and X. Qi, Appl. Phys. A: Mater. Sci. Process., 2024, 130, 612.
44. X. Shen, X. Lin, Y. Peng, Y. Zhang, F. Long, Q. Han, Y. Wang and L. Han, Nano-Micro Lett., 2024, 16, 201.
45. J. Huang, M. Wu, J. Yan, X. Zhu, P. Hu, J. Sun, Q. Li, Z. Lei, Z.-H. Liu and X. He, ACS Nano, 2025, 19, 26600–26611.
46. B. Mukherjee, A. Zulkefli, K. Watanabe, T. Taniguchi, Y. Wakayama and S. Nakaharai, Adv. Funct. Mater., 2020, 30, 2001688.
47. S. Xu, C. Kang, Z. Huang, Z. Zhang, H. Rao, Z. Pan and X. Zhong, Sol. RRL, 2022, 6, 2100989.
48. H. Zhang, Y. Li, S. Tan, Z. Chen, K. Song, S. Huang, J. Shi, Y. Luo, D. Li and Q. Meng, J. Colloid Interface Sci., 2022, 608, 3151–3158.
49. X. Li, N. Fu, A. Mei, X. Peng, H. Wang, Y. Lin and J. Du, Sol. RRL, 2024, 8, 2400295.
50. A. Mei, X. Peng, X. Li, G. Zhang, S. Lin and N. Fu, ACS Appl. Mater. Interfaces, 2024, 16, 31209–31217.
51. T. Xu, X. Zhang, F. Liu, H. Guo, J. Zhang, S. Cai, D. Li, Y. Zhang, Y. Guan and W. Yu, Mater. Horiz., 2025, 12, 3436–3443.
52. H. Lin, Z. Li and N. Fu, ACS Appl. Energy Mater., 2025, 8, 14764–14772.
53. L. Shen, P. Song, L. Zheng, L. Wang, X. Zhang, K. Liu, Y. Liang, W. Tian, Y. Luo, J. Qiu, C. Tian, L. Xie and Z. Wei, Adv. Mater., 2023, 35, e2301624.
54. H. Zhong, W. Li, Y. Huang, D. Cao, C. Zhang, H. Bao, Z. Guo, L. Wan, X. Zhang, X. Zhang, Y. Li, X. Ren, X. Wang, D. Eder, K. Wang, S. F. Liu and S. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 5183–5193.
55. H. Guo, Y. Fang, H.-B. Cheng, J. Wu, Y. Lei, S. Wang, X. Li, Y. Dai, W. Xiang, D.-J. Xue, Y. Lin and A. Hagfeldt, Angew. Chem., Int. Ed., 2022, 61, e202204148.
56. D. Wu, H. Zhou, X. Lai, X. Liu, K. Sang, Y. Chen, M. Chen, J. Wei, S. Wu and Q. Pang, Small, 2024, 20, 2308783.
57. J. Liang, Z. Liu, L. Qiu, Z. Hawash, L. Meng, Z. Wu, Y. Jiang, L. K. Ono and Y. Qi, Adv. Energy Mater., 2018, 8, 1800504.
58. T. Markvart, Appl. Phy. Lett., 2007, 91, 064102.
59. M. Stolterfoht, P. Caprioglio, C. M. Wolff, J. A. Márquez, J. Nordmann, S. Zhang, D. Rothhardt, U. Hörmann, Y. Amir, A. Redinger, L. Kegelmann, F. Zu, S. Albrecht, N. Koch, T. Kirchartz, M. Saliba, T. Unold and D. Neher, Energy Environ. Sci., 2019, 12, 2778–2788.
60. J. Warby, S. Shah, J. Thiesbrummel, E. Gutierrez-Partida, H. Lai, B. Alebachew, M. Grischek, F. Yang, F. Lang, S. Albrecht, F. Fu, D. Neher and M. Stolterfoht, Adv. Energy Mater., 2023, 13, 2303135.

---

Footnote

† Yi-Ping Zhang (Y.-P. Zhang) and Xinwei Li (X. Li) contributed equally to this work.

---