Phase-pure Ruddlesden–Popper tin halide perovskites for solar energy conversion applications

Abstract

Two-dimensional Ruddlesden–Popper (RP) tin halide perovskites have recently shown promise in solar energy conversion applications owing to their low-toxicity and low-cost processability. However, RP tin halide perovskites typically consist of multiple phases with enormously disordered crystal orientation, lowering their power conversion efficiency. Herein, we show a new method to prepare phase-pure RP tin hybrid perovskite films by fluorinating the conventionally used spacer cation C₆H₅CH₂NH₃⁺ (abbreviated as FBE⁺). This solvophobic engineering strategy decreases the interaction among spacer cations and solvent molecules to induce the self-assembly of FBE⁺ cations into micelles. These micelles in the solvent capture tin iodide components, forming crystal nuclei. This “simultaneous” crystallization mode leads to uniformly distributed nuclei and further results in a phase-pure RP perovskite with vertical crystal orientation. The resultant phase pure FBE₂FA₃Sn₄I₁₃ perovskite exhibits an improved carrier mobility and much reduced trap density, which are favourable for solar energy conversion applications. A power conversion efficiency of 9.25% was achieved for the FBE₂FA₃Sn₄I₁₃ perovskite solar cell, which is among the highest efficiencies reported for state-of-the-art two-dimensional RP tin hybrid perovskite cells.

---

Introduction

Tin halide perovskites have attracted considerable attention for solar energy conversion applications owing to their low toxicity and adjustable band gap.^1–3 Preventing the oxidation of Sn²⁺ cations in tin halide perovskites is challenging, which leads to abundant Sn vacancies and thus limits their performance in optoelectronic devices.^4–6 Several strategies such as additive engineering and pseudo-halide anion engineering have been proposed to improve the photovoltaic performance of tin perovskite-based devices.^7–12 Recently, perovskite solar cells (PSCs) using a 2D/3D heterogeneous FASnI₃ absorber obtained by substituting FAI with FPEABr have reached an impressive power conversion efficiency (PCE) of 14.8%.¹³

Two-dimensional Ruddlesden–Popper (RP) tin hybrid perovskites can be descried with a general formula of R₂A_n−1Sn_nI_3n+1, where the integer n represents the number of [SnI₆] octahedral sheets separated by bulky organic cations (R), and A is HC(NH₂)₂⁺ (FA⁺), CH₃NH₃⁺ (MA⁺) or Cs⁺.^14–18 Due to the difference in dielectric constants for the organic and inorganic layers, RP perovskites naturally form a quantum-well structure,¹⁹ which is helpful for preventing Sn²⁺ oxidization.^20,21 However, carrier transport within the RP perovskite thin films might be hindered by the less conductive organic layer. Therefore, for ideal cases the perpendicular orientation of RP perovskite phases needs to form pathways for rapid charge transfer. Low-temperature solution-processed RP perovskites usually are multi-phases (n = 1, 2, 3, etc.) containing compounds induced by super-saturation between the surface and interior of the precursor.^22–24 Detailed investigation demonstrates a spatial distribution of concentration for these chemical components from the solution-air interface to the bulk solution.^25–27 Such a “top-down” crystallization leads to a RP perovskite film featuring a random distribution of different single RP phases.

The energy barrier of charge-transfer in phase-pure RP perovskites could be reduced compared to a multiphase counterpart.^28–31 For example, Lin et al. introduced a carboxylic acid additive to increase the discrepancy in the activation energy of the nucleation process of 2D phases, resulting in RP perovskite films with a specific n integer.²⁸ An increase in photoluminescence energy and photo-stability was observed in these films. Likewise, Li et al. added a high-boiling point ionic liquid solvent n-butylammoniumacetate in precursor solution to coordinate with SnI₂.³⁰ They found that the method could delay the precipitation of tin iodide composition and reduce the speed of crystallization, eventually forming phase-pure RP perovskites. These reports verify the critical role of regulating the RP perovskite crystallization process in the formation of phase-pure perovskites. However, residual impurities might be present in final films due to adding non-raw materials in perovskite precursor solutions.

Solvophobic effect is widely observed for solute molecule self-assembly due to the repulsion among solutes and solvents.³² The formed aggregates in solvents first reach up to a critical nucleus size followed by growth of particles, which can probably break the “top-down” crystallization mode of RP perovskites. Here, we show a method based on solvophobic effect to deposit phase-pure RP tin halide perovskite films for solar energy conversion applications. We first functionalized phenyl components by substitution the hydrogen on 2 and 4-position in the prototypical cation C₆H₅CH₂NH₃⁺ (BE⁺) with fluoro-group (FBE⁺). We further revealed that FBE⁺ is inclined to self-assemble into micelles in the solution. This essentially induces uniformly distributed nuclei composed of inorganic sheets and large organic cations for phase-pure RP perovskite films. The resultant FBE₂FA₃Sn₄I₁₃ PSC device achieved an impressive power conversion efficiency of 9.25%, which is among the highest efficiencies of the reported state-of-the-art cells using 2D RP tin perovskites.

Results and discussion

Here, we developed an organic cation with two characteristics for RP tin halide perovskites. Firstly, the cation is highly solvophobic, which could lead to micelle formation at a low concentration. Secondly, the cation has a strong interaction with tin iodide components for micelle nucleation. Keeping these in mind, we introduced fluorine to substitute the hydrogen at the ortho- and para-positions of the phenyl ring in C₆H₅CH₂NH₃⁺ (BE⁺), referred to as FBE⁺. As the solvent evaporates, FBE⁺ self-assembles into micelles. These micelles capture tin iodide components and form crystal nuclei composed of inorganic sheets and large organic cations, which leads to a phase-pure and vertically oriented RP perovskite film (Fig. 1a).

Fig. 1 (a) Schematic of proposed scenarios of the growth of phase-pure tin-based perovskite films treated by solvophobic engineering. Molecular electrostatic potential surface distributions of (b) BE and (c) FBE. (d) Interfacial tension measurements with the solvent as a function of spacer molecule concentration. ¹H NMR signals measured for (e) BEI and compounds of BEI and SnI₂, and (f) FBEI and compounds of FBEI and SnI₂ in deuterated DMSO solution.

The electrostatic potential (ESP) distribution on the van der Waals (vdW) surface for BE and FBE in DMF was first evaluated with density functional theory (DFT) calculations. Fig. 1b shows that for the BE, the distribution of charge around the terminal benzyl is quite symmetric in the range from 10.14 to 15.58 kcal mol⁻¹, while that around the difluorinated benzyl of FBE exhibits significant asymmetry, with a negative field surrounding the F atom (−45.51 and −17.74 kcal mol⁻¹) and a positive field enveloping the –CH moiety (21.50 and 16.62 kcal mol⁻¹) of the terminal benzyl (Fig. 1c). This offers a 2,4-difluoropheny dipolar. Fig. S1† compares the percentage of each ESP range to the whole molecular surface. Compared to the BE, the FBE shows a much wider ESP range. Therefore, the FBE possesses a more significant fluctuated ESP distribution on the vdW surface than BE.

The orientation change of solvent molecules is a key parameter to assess the solvation strength of the solute. Fig. S2a† shows the ensemble-averaged orientation of DMF in the first solvation shell of BE⁺ and FBE⁺, where θ is defined as the angle between the direction of a C=O bond and the C1 axis of the terminal benzyl of the spacer cation passing through the DMF–O. The value of θ close to 0° indicates a perfectly H-bonded configuration, while a θ value higher than 90° implies a non-H-bonded configuration. Compared to that for BE⁺ (43°), the θ-value of DMF for FBE⁺ (85°) is close to 90°, suggesting a weaker interaction between the solvent molecule and the difluorination due to the change in the electric potential and dipolar character of the spacer (Fig. S2b†).³³ The solvophobicity facilitates the formation of micelles.

The critical micelle concentration (CMC) is a key indicator of the interaction between spacer cations and solvents, which could be determined experimentally using the interfacial tension characterization. As shown in Fig. 1d, the interfacial tension for both samples first decreases and then increases with increasing the spacer cation concentration. The concentration at the turn-over point represents the CMC for each sample. The CMC of FBE and BE is ∼0.06 and ∼0.19 mM mL⁻¹, respectively, far below the concentration of FBE and close to that of BE in precursor solution (0.2 mM mL⁻¹). This indicates that micelles could be formed in the precursor solution with a lower concentration of FBE. Micelles could also be nucleation sites, changing the “top-down” crystallization mode.

We further investigated the interaction among components in the precursor solution using ¹H nuclear magnetic resonance (NMR) spectroscopy measurements. As shown in Fig. 1e and f, the chemical shift for the proton signal from the benzene ring (the peak marked with orange circles for example) from BEI moves from 7.41 ppm to 7.45 ppm, and that from the FBEI shifts to 7.54 ppm from 7.61 ppm when adding SnI₂ into the deuterated DMSO solution containing the spacer cations. A larger change of the chemical shift (0.04 ppm vs. 0.07 ppm) indicates a stronger interaction between FBE⁺ and tin iodide compounds. Combining the result of the ESP distribution in spacer molecules (Fig. S2b†), the strong intermolecular interaction probably results from a much greater degree of amino group protonation because the highly electronegative fluorine leads to the inductive effect, the shifting of electrons in the amino group to the benzene ring. The strong interaction facilitates micelles to capture tin iodide components, forming crystal nuclei.

The residual solvent in the BE₂FA₃Sn₄I₁₃ and FBE₂FA₃Sn₄I₁₃ films before annealing was quickly extracted by using a molecular pump. Fig. 2a presents the scanning electron microscopy (SEM) image of the BE₂FA₃Sn₄I₁₃ sample, from which nuclei with different sizes and shapes on the PEDOT:PSS film surface can be clearly observed. The thin flake-like domains (marked by a pink dotted line) in small quantities probably correspond to low-n phases and the blocky ones (marked by a blue dotted line) in bulk to high-n phases.³⁴ When using FBE⁺ as the spacer cation, numerous and regular nuclei corresponding to high-n phases uniformly disperse over the entire surface of the FBE₂FA₃Sn₄I₁₃ film (Fig. 2b). The density of nuclei was estimated to be ∼2.81 × 10⁷ mm⁻² for the FBE₂FA₃Sn₄I₁₃ film, being twice higher than that of the BE₂FA₃Sn₄I₁₃ film (∼1.45 × 10⁷ mm⁻²) based on the ratio of surface coverage to average domain area.^35–37 The cross-sectional SEM image of the BE₂FA₃Sn₄I₁₃ film in Fig. 2c shows diverse sizes and shapes of nuclei in different heights, while that of the FBE₂FA₃Sn₄I₁₃ film in Fig. 2d exhibits a similar morphology of nuclei, which basically coincides with top-view SEM images. This probably results from a “simultaneous” crystallization in FBE-based perovskite solution.

Fig. 2 The top-view SEM images and the cross-sectional SEM images of the BE₂FA₃Sn₄I₁₃ (a and c) and FBE₂FA₃Sn₄I₁₃ (b and d) films. The “flake-like” domains marked by a pink dotted line correspond to low-n phases and the blocky domains marked by a blue dotted line correspond to high-n phases. (e) Steady-state PL spectra of the BE₂FA₃Sn₄I₁₃ and FBE₂FA₃Sn₄I₁₃-based films from the substrate side. (f) XRD patterns of perovskite films.

Photoluminescence (PL) measurements from the perovskite side (front excitation) and the substrate side (back excitation) were performed to probe the phase distribution. In the case of front-excitation, the samples showed an emission peak centered at around 1.54 eV (Fig. S3†). In the case of back-excitation, multiple emission peaks were identified at ∼1.99 and ∼1.78 eV for the BE₂FA₃Sn₄I₁₃-based film, corresponding to the RP phases with n = 1 and 2, respectively (Fig. 2e). The difference in the PL spectra between front- and back-excitations suggests that the BE₂FA₃Sn₄I₁₃ film contains multiple phases, in which the low-n phases majorly locate near the substrate side, and the high-n phase at the upper surface.^25–27 The PL spectra of the FBE₂FA₃Sn₄I₁₃-based film exhibited a clear single peak at ∼1.54 eV corresponding to a high-n phase (Fig. 2e). The signals associated with the low-n phases are almost negligible. The result reveals that the single phase of the RP film is deposited.

As shown in Fig. 2f, both samples exhibited similar X-ray diffraction (XRD) peaks at ∼14.0° and ∼28.2°, which correspond to the (111) and (202) planes of the RP perovskite crystal structure, respectively, aligning with the powder XRD peaks of the BE₂FA₃Sn₄I₁₃ FBE₂FA₃Sn₄I₁₃ single crystals (Fig. S4a and b†). The appearance of diffraction peaks for 2θ ≤ 10° in the BE₂FA₃Sn₄I₁₃ film indicates the existence of low-n value phases, in which the XRD peaks at ∼4.1° and ∼5.4° can be ascribed to the (200) planes in the n = 2 phase and the (200) in the n = 1 phase, respectively. This result is consistent with the powder XRD peaks of the BE₂FASn₂I₇ and BE₂SnI₄ single crystals, respectively (Fig. S4c and d†). By contrast, the disappeared diffraction peaks (2θ ≤ 10°) for the FBE₂FA₃Sn₄I₁₃ film can be ascribed to the fluorinated cation on narrow-phase distribution for RP perovskites. The FBE₂FA₃Sn₄I₁₃ film shows a higher intensity of the diffraction peak and a smaller full width at half maximum (0.25°) of the (111) plane, indicating its high crystallinity. This can be correlated with a strong interlayer molecular interaction between F atoms and their adjacent phenyl rings align. ^5,38

We suggest that the crystallization of BE₂FA₃Sn₄I₁₃ and FBE₂FA₃Sn₄I₁₃ perovskites is different (Fig. 3 and Video S1†). As both samples have a basic chemical structure and preparation process in common, the only difference in the end group of spacer cations is probably the root cause of their crystallization behaviours. For the BE₂FA₃Sn₄I₁₃ sample, the precursor solution at the solution–air surface first reaches the super-saturation and releases tin iodide layers, with which FA⁺ forms complexes to nucleate and grow grains. BE⁺ is inclined to remain in solution due to its high solubility, corresponding to a higher CMC value as shown in Fig. 1d. The “top-down” crystallization mode results in a gradient distribution of phases in the RP perovskite film, reflected by small-n phases close to the substrate and large-n phases to the electrode (Fig. 3a). In contrast, due to a solvophobic interaction between difluorinated benzyl and solvents, FBE⁺ cations self-assemble into micelles in the bulk solvent. The micelles achieve a state of super-saturation, resulting in spatial uniform distribution of crystal nuclei. The nuclei then initiate the growth of RP perovskite grains, leading to a homogeneous and narrow n-value distribution (Fig. 3b).

Fig. 3 Schematic illustration of growth kinetics and phase distribution of (a) BE₂FA₃Sn₄I₁₃- and (b) FBE₂FA₃Sn₄I₁₃-based perovskites. The micelle is marked by a black dotted line.

Grazing incidence wide-angle X-ray scattering (GIWAXS) measurements were performed to elucidate the crystal orientation. The presence of Debye–Scherrer rings at low q values in Fig. 4a can be ascribed to the (k00) planes of the n = 1 and n = 2 phases of the BE₂FA₃Sn₄I₁₃ film, which is consistent with other reports. ^24,39–41 The intensity of the Debye–Scherrer rings corresponding to the (111) and (202) planes of the high-n phases presents an isotropic distribution, indicating considerable randomness in the orientation of the crystal grains. The sharp and discrete Bragg spots corresponding to the (111) and (202) planes were observed in the GIWAXS pattern of the FBE₂FA₃Sn₄I₁₃ sample (Fig. 4b). This suggests that crystal grains are oriented preferably vertical to the substrate.^42,43

Fig. 4 GIWAXS patterns of (a) BE₂FA₃Sn₄I₁₃ and (b) FBE₂FA₃Sn₄I₁₃ films. Dark J–V measurements of (c) electron-only and (d) hole-only devices based on the BE₂FA₃Sn₄I₁₃ and FBE₂FA₃Sn₄I₁₃-based perovskites. (e) Photocurrent J–V curves and (f) EQE spectra and integrated current of the BE₂FA₃Sn₄I₁₃ and FBE₂FA₃Sn₄I₁₃-based devices.

We further evaluated the trap density in the RP perovskite films by using electron-only and hole-only devices with the structure of ITO/SnO₂/perovskite/PCBM/BCP/Au (Fig. 4c) and ITO/PEDOT:PSS/perovskite/spiro-OMeTAD/Au (Fig. 4d), respectively. The electron and hole trap density in the BE₂FA₃Sn₄I₁₃-based perovskite are up to a high value of 1.05 × 10¹⁶ and 1.32 × 10¹⁶ cm⁻³, respectively, as compared to only 2.24 × 10¹⁵ and 2.40 × 10¹⁵ cm⁻³, respectively, in the FBE₂FA₃Sn₄I₁₃-based perovskite. The reduced defects could be attributed to the suppression of small n-value phases acting as carrier traps in the FBE₂FA₃Sn₄I₁₃-based perovskite. The carrier mobility μ of RP perovskites can further be obtained by the Mott–Gurney equation as follows:⁴⁴

where J is the current density, ε₀ is the vacuum permittivity, ε_r is the vacuum permittivity, V is the base voltage and d is the thickness of the perovskite film. The electron and hole mobility of the FBE₂FA₃Sn₄I₁₃-based film are calculated to be 1.95 × 10⁻³ and 1.43 × 10⁻³ cm² V⁻¹ s⁻¹, respectively, both of which increase by one and two orders of magnitude compared to those of the BE₂FA₃Sn₄I₁₃ film (1.48 × 10⁻⁴ and 1.55 × 10⁻⁵ cm² V⁻¹ s⁻¹, respectively). The superior charge transport property of the FBE₂FA₃Sn₄I₁₃-based perovskite is mainly attributed to the pure phase with the vertical orientation, which guarantees efficient carrier transport.

Next, we assessed the photovoltaic performance of the RP PSCs in the state-of-the-art planar architecture of ITO/PEDOT:PSS/perovskite/PCBM/BCP/Ag. The photovoltaic performances of the devices were measured under 1-sun illumination. Both the BE₂FA₃Sn₄I₁₃-based and FBE₂FA₃Sn₄I₁₃-based devices exhibit great reproducibility with an average PCE of 3.10% and 8.68%, respectively (Fig. S6†). Fig. 4e depicts the current density–voltage (J–V) characteristics of champion devices under 1-sun illumination, and the corresponding photovoltaic parameters are tabulated in Table S1.† The FBE₂FA₃Sn₄I₁₃-based device shows a high PCE of 9.25% with an open-circuit voltage (V_oc) of 0.69 V, a short circuit current density (J_sc) of 21.10 mA cm⁻², and a fill factor (FF) of 63.5%, while the control device displays a relatively low PCE of 4.13%, V_oc of 0.56 V, J_sc of 15.17 mA cm⁻² and FF of 48.6%. The performance improvement is mainly attributed to the enhanced J_sc in the FBE₂FA₃Sn₄I₁₃-based device, which is a consequence of the highly vertically oriented RP perovskite with a high charge mobility. The integrated J_sc values from corresponding external quantum efficiency (EQE) were calculated to be 14.78 and 20.67 mA cm⁻² for the BA₂FA₃Sn₄I₁₃-based and FBA₂FA₃Sn₄I₁₃-based devices, respectively, which are consistent with the measured J_sc values from the PSCs (Fig. 4f).

We further extracted the Urbach energy (E_u) from the fitted semi-log plot of EQE values versus photon energy in the low-energy regime to quantify the degree of energy disorder (Fig. S7†). ^45,46 The E_u was calculated to be 29.4 and 21.1 meV for the BE₂FA₃Sn₄I₁₃- and FBE₂FA₃Sn₄I₁₃-based films, respectively, suggesting lower structural disorder and better crystalline quality by fluorination of the cation spacer. Such an ordered structure of the FBE₂FA₃Sn₄I₁₃-based perovskite could reduce carrier recombination and promote carrier transportation, which is essential to obtain high V_oc and J_sc.

To investigate carrier transportation and recombination in the RP perovskites, we conducted transient photovoltage/photocurrent decay measurements (TPV/TPC). The solar cells were kept under open-circuit conditions under illumination, and subsequently were excited with a green LED pulse (2 μs) that generated a small perturbation of the V_oc (ΔV). The carriers photo-generated by the laser, proportional to ΔV, were forced to recombine, leading to a transient decay of photovoltage. With increasing the light intensity, a faster decay was registered due to the higher photo-generated charge carriers (Fig. S8a and b†). As the voltage decay rate describes the change in carrier density versus time, carrier recombination lifetimes (τ_re) could be obtained by fitting the TPV curves. Compared to the control device, the FBE₂FA₃Sn₄I₁₃-based device exhibits longer electron and hole lifetimes, indicating the reduction of carrier recombination due to fewer film defects (Fig. 5a). TPC decays were measured under the same conditions as TPV but were held in short circuit conditions (i.e., assuming that all photo-generated carriers are collected). The downward trend in circuit current (ΔJ) when the light intensity is increased is similar to that in ΔV for TPV (Fig. S8c and d†). Fig. 5b shows the carrier transport lifetimes (τ_tr) extracted from TPC measurements. At the same voltage, the τ_tr of both electrons and holes for the FBE₂FA₃Sn₄I₁₃-based device is shorter than that for the BE₂FA₃Sn₄I₁₃-based device, verifying that carrier collection is more efficient in the former.

Fig. 5 (a) Charge recombination time, (b) transport time, (c) calculated electron diffusion length, and (d) hole diffusion length of the BE₂FA₃Sn₄I₁₃-based and FBE₂FA₃Sn₄I₁₃-based devices.

Carrier diffusion length (L) could be calculated by using eqn (2): ^40,41

where d is the photoactive layer thickness, and c is a constant linked to the charge conduction mechanism. The electron and hole diffusion lengths at 100 mW cm⁻² light bias increase from 460 to 764 nm and from 356 to 725 nm, respectively, when BE⁺ as the spacer cation is replaced by FBE⁺. This corresponds to an effective charge collection in the optimum PSC (Fig. 5c and d), which leads to a higher J_sc.

The differential capacitance C is defined as the ability of a device to store charges, which could be quantified by^47,48

where S is the active area, ΔV is the amplitude of relative TPV signals, and Δq is the amount of charges calculated by the integration of TPC decay. Fig. S7† shows that the capacitance in both samples increases exponentially with V_oc. In general, this exponential capacitance is referred to as the chemical capacitance of the perovskite active layer, which is highly correlated with the electron trapping–detrapping procedure through trap states. The FBE₂FA₃Sn₄I₁₃-based device presents a smaller capacitance, indicating a decrease of charge accumulation within traps.

Upon exposure to ambient air with about 15% relative humidity and 20 °C for 24 hours, the unencapsulated FBE₂FA₃Sn₄I₁₃-based device retains 91% of the initial efficiency, whereas the PCE of the BE₂FA₃Sn₄I₁₃ device decreases by 50% as illustrated in Fig. S9.† The improved ambient stability results from the high-quality FBE₂FA₃Sn₄I₁₃ film, which is capable of resisting the erosion caused by water and oxygen molecules to a large extent.

Conclusions

In summary, we propose the fluorination of spacer cations for achieving phase-pure tin hybrid RP perovskites with vertically aligned grains. By introducing an amphiphilic salt spacer (FBEI) instead of the traditional halide spacers (BEI), we successfully demonstrated that the solvophobic effect could offer an opportunity for obtaining a uniformly spatial distribution of perovskite nuclei by generating micelles in the bulk of the perovskite precursor. The FBE₂FA₃Sn₄I₁₃-based device exhibited a high PCE of 9.25% and air stability. This work provides an effective way to develop efficient and stable RP tin-based perovskites.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

H. P. conducted the device fabrication and prepared the manuscript. H. P., Y. W., S. G., N. C. and J. Z. synthesized the materials and measured the device performance. Y. Z. and. S. C., X. G performed the calculations. W. Y., Y. X. and Y. S. revised the manuscript. M. W. guided and supervised the whole study.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge financial support from Research Foundation of Yangtze University (No. 8021003201-2023) and the Science Foundation of Educational Commission of Hubei Province of China (No. T2020008), the Natural Science Foundation of Hubei Province of China (No. 2023AFA025), the Young Fund of the Natural Science Foundation of Hubei Province of China (No. 2024AFB430), the Young Talent Support Project of Henan Province (2024HYTP001), and the National Natural Science Foundation of China (No. 12375008, 12075036 and 12304473).

Notes and references

1. R. Milot, M. Klug, C. Davies, Z. Wang, H. Kraus, H. Snaith, M. Johnston and L. Herz, Adv. Mater., 2018, 30, e1804506.
2. A. Abate, ACS Energy Lett., 2023, 8, 1896–1899.
3. Y. Li, H. Zhou, M. Xia, H. Shen, T. Wang, H. Gao, X. Sheng, Y. Han, Z. Chen, L. Dou, H. Zhu and E. Shi, Sci. Adv., 2023, 9, eadh0517.
4. L. Dai, I. C. P. Roca, H. Ban, Z. Zhang, Q. Sun, X. Li, A. Gu, W. Yang, H. Yu, Y. Shen and M. Wang, Small, 2023, 19, e2208062.
5. Z. Zhang, L. Dai, M. Zhang, H. Ban, Z. Liu, H. Yu, A. Gu, X. L. Zhang, S. Chen, Y. Wang, Y. Shen and M. Wang, ACS Appl. Mater. Interfaces, 2023, 15, 36594–36601.
6. T. Macdonald, L. Lanzetta, X. Liang, D. Ding and S. Haque, Adv. Mater., 2023, 35, e2206684.
7. X. Jiang, H. Li, Q. Zhou, Q. Wei, M. Wei, L. Jiang, Z. Wang, Z. Peng, F. Wang, Z. Zang, K. Xu, Y. Hou, S. Teale, W. Zhou, R. Si, X. Gao, E. Sargent and Z. Ning, J. Am. Chem. Soc., 2021, 143, 10970–10976.
8. W. Shen, H. Cai, Y. Kong, W. Dong, C. Bai, G. Liang, W. Li, J. Zhao, F. Huang, Y. Cheng and J. Zhong, Small, 2023, 19, e2302194.
9. C. Kuan, J. Chih, Y. Chen, B. Liu, C. Wang, C. Hou, J. Shyue and E. Diau, ACS Energy Lett., 2022, 7, 4436–4442.
10. H. Jia, H. Shi, R. Yu, H. Ma, Z. Wang, C. Zou and Z. Tan, Small, 2022, 18, e2200036.
11. H. Li, Y. Xu, S. Ramakrishnan, Y. Zhang, M. Cotlet, T. L. Xu and Q. Yu, Cell Rep. Phys. Sci., 2022, 3, 101060.
12. M. Li, W. Zuo, Y. Yang, M. Aldamasy, Q. Wang, S. Cruz, S. Feng, M. Saliba, Z. Wang and A. Abate, ACS Energy Lett., 2020, 5, 1923–1929.
13. B. Yu, Z. Chen, Y. Zhu, Y. Wang, B. Han, G. Chen, X. Zhang, Z. Du and Z. He, Adv. Mater., 2021, 33, e2102055.
14. J. Zhao, Z. Zhang, G. Li, M. Aldamasy, M. Li and A. Abate, Adv. Energy Mater., 2023, 13, 2204233.
15. Y. Reo, T. Choi, J. Go, S. Jeon, B. Lim, H. Zhu, A. Liu and Y. Noh, ACS Energy Lett., 2023, 8, 3088–3094.
16. H. Gu, J. M. Xia, C. Liang, Y. Chen, W. Huang and G. Xing, Nat. Rev. Mater., 2023, 8, 533–551.
17. C. Wang, S. Cui, Y. Ju, Y. Chen, S. Chang and H. Zhong, Adv. Funct. Mater., 2023, 33, 2301304.
18. Y. Chen, Z. Wang, Y. Wei, Y. Liu and M. Hong, Angew. Chem., Int. Ed., 2023, 62, e2023016.
19. D. Mitzi, Inorg. Chem., 2000, 39, 6107–6113.
20. Y. Liao, H. Liu, W. Zhou, D. Yang, Y. Shang, Z. Shi, B. Li, X. Jiang, L. Zhang, L. N. Quan, R. Quintero-Bermudez, B. Sutherland, Q. Mi, E. Sargent and Z. Ning, J. Am. Chem. Soc., 2017, 139, 6693–6699.
21. T. Zhang, Q. Sun, X. L. Zhang, Y. Shen and M. K. Wang, APL Mater., 2021, 9, 020906.
22. C. Sun, Y. Jiang, M. Cui, L. Qiao, J. Wei, Y. Huang, L. Zhang, T. He, S. Li, H. Hsu, C. Qin, R. Long and M. Yuan, Nat. Commun., 2021, 12, 2207.
23. R. Quintero-Bermudez, A. Gold-Parker, A. Proppe, R. Munir, Z. Yang, S. Kelley, A. Amassian, M. Toney and E. Sargent, Nat. Mater., 2018, 17, 900–907.
24. C. Liang, H. Gu, Y. D. Xia, Z. Wang, X. Liu, J. M. Xia, S. Zuo, Y. Hu, X. Gao, W. Hui, L. Chao, T. Niu, M. Fang, H. Lu, H. Dong, H. Yu, S. Chen, X. Ran, L. Song, B. Li, J. Zhang, Y. Peng, G. Shao, J. Wang, Y. Chen, G. Xing and W. Huang, Nat. Energy, 2021, 6, 38–45.
25. Q. Cao, P. Li, W. Chen, S. Zang, L. Han, Y. Zhang and Y. Song, Nano Today, 2022, 43, 101394.
26. N. Zhou, Y. Shen, L. Li, S. Tan, N. Liu, G. Zheng, Q. Chen and H. Zhou, J. Am. Chem. Soc., 2018, 140, 459–465.
27. P. Liu, N. Han, W. Wang, R. Ran, W. Zhou and Z. Shao, Adv. Mater., 2021, 33, e2002582.
28. J. Lin, D. Chen, C. Wu, C. Hsu, C. Chien, H. Chen, P. Chou and C. Chiu, Angew. Chem., Int. Ed., 2021, 60, 7866–7872.
29. C. Liang, H. Gu, J. Xia, T. Liu, S. Mei, N. Zhang, Y. Chen and G. Xing, Carbon Energy, 2023, 5, e251.
30. G. Li, Z. Su, M. Li, F. Yang, M. H. Aldamasy, J. Pascual, F. Yang, H. Liu, W. Zuo, D. Di Girolamo, Z. Iqbal, G. Nasti, A. Dallmann, X. Gao, Z. Wang, M. Saliba and A. Abate, Adv. Energy Mater., 2021, 11, 2101539.
31. S. Sidhik, Y. Wang, M. De Siena, R. Asadpour, A. Torma, T. Terlier, K. Ho, W. Li, A. Puthirath, X. Shuai, A. Agrawal, B. Traore, M. Jones, R. Giridharagopal, P. Ajayan, J. Strzalka, D. Ginger, C. Katan, M. Alam, J. Even, M. Kanatzidis and A. Mohite, Science, 2022, 377, 1425–1430.
32. T. Greaves and C. Drummond, Chem. Soc. Rev., 2013, 42, 1096–1120.
33. S. Roy, B. Biswas, N. Ghosh, P. Singh and J. Mondal, J. Phys. Chem. C, 2019, 123, 27012–27019.
34. P. Li, Y. Zhang, C. Liang, G. Xing, X. Liu, F. Li, X. Liu, X. Hu, G. Shao and Y. Song, Adv. Mater., 2018, 30, 1805323.
35. Z. Lin, Y. Zhao, C. Zhou, R. Zhong, X. Wang, Y. Tsang and Y. Chai, Sci. Rep., 2015, 5, 18596.
36. C. Xu, Z. Liu and E. Lee, J. Mater. Chem. C, 2020, 8, 15860.
37. S. Chu, W. Chen, Z. Fang, X. Xiao, Y. Liu, J. Chen, J. Huang and Z. Xiao, Nat. Commun., 2021, 12, 147.
38. H. Ren, S. Yu, L. Chao, Y. Xia, Y. Sun, S. Zuo, F. Li, T. Niu, Y. Yang, H. Ju, B. Li, H. Du, X. Gao, J. Zhang, J. Wang, L. Zhang, Y. Chen and W. Huang, Nat. Photonics, 2020, 14, 154–163.
39. J. Qing, X. Liu, M. Li, F. Liu, Z. Yuan, E. Tiukalova, Z. Yan, M. Duchamp, S. Chen, Y. Wang, S. Bai, J. Liu, H. Snaith, C. Lee, T. Sum and F. Gao, Adv. Energy Mater., 2018, 8, 1800185.
40. Y. Yang, C. Liu, H. Kanda, Y. Ding, H. Huang, H. Chen, B. Ding, Y. Liang, X. Liu, M. Cai, P. Dyson, S. Dai and M. Nazeeruddin, Adv. Funct. Mater., 2021, 31, 2104868.
41. S. Sidhik, W. Li, M. Samani, H. Zhang, Y. Wang, J. Hoffman, A. Fehr, M. Wong, C. Katan, J. Even, A. Marciel, M. Kanatzidis, J. Blancon and A. Mohite, Adv. Mater., 2021, 33, 2007176.
42. Y. Zhang, M. Chen, T. He, H. Chen, Z. Zhang, H. Wang, H. Lu, Q. Ling, Z. Hu, Y. Liu, Y. Chen and G. Long, Adv. Mater., 2023, 35, e2210836.
43. Q. Fu, M. Chen, Q. Li, H. Liu, R. Wang and Y. Liu, J. Am. Chem. Soc., 2023, 145, 21687–21695.
44. H. Li, X. Wang, T. Zhang, X. Gong, Q. Sun, H. Pan, Y. Shen, S. Ahmad and M. Wang, Adv. Funct. Mater., 2019, 29, 1903293.
45. Q. Dong, C. Zhu, M. Chen, C. Jiang, J. Guo, Y. Feng, Z. Dai, S. Yadavalli, M. Hu, X. Cao, Y. Li, Y. Huang, Z. Liu, Y. Shi, L. Wang, N. Padture and Y. Zhou, Nat. Commun., 2021, 12, 973.
46. W. Zhang, Y. Xie, D. Xiong, X. Zeng, Z. Li, M. Wang, Y. Cheng, W. Chen, K. Yan and S. Yang, ACS Appl. Mater. Interfaces, 2014, 6, 9698–9704.
47. X. Zhang, S. Wang, W. Zhu, Z. Cao, A. Wang and F. Hao, Adv. Funct. Mater., 2022, 32, 2108832.
48. T. Du, J. Kim, J. Ngiam, S. Xu, P. Barnes, J. Durrant and M. McLachlan, Adv. Funct. Mater., 2018, 28, 1801808.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta02405b

---