Coupling of MAPbI₃ microcrystals with conductive polyaniline for efficient visible-light-driven H₂ evolution

Abstract

Although the methylammonium lead iodide (MAPbI₃) hybrid perovskite has shown great potential in photocatalytic H₂ evolution, its performance is significantly suppressed due to the insufficient charge separation. Herein, a strongly coupled 3D/1D composite photocatalyst of MAPbI₃ microcrystals with conductive polyaniline nanowires (PANI NWs) is developed for efficient photocatalytic H₂ evolution in MAPbI₃-saturated HI aqueous solution under visible light irradiation. The most active MAPbI₃ (100 mg)–PANI NW (7 mg) composite exhibits a H₂ evolution rate of 38.8 μmol h⁻¹, which is 29 times higher than that of pristine MAPbI₃. Spectroscopic and photoelectrochemical measurements reveal that the coupled PANI NWs can greatly improve the charge separation, thereby considerably enhancing the photocatalytic H₂ evolution of the composite photocatalyst.

---

Photocatalytic H₂ evolution represents a highly promising route to effectively convert solar energy to clean H₂ energy.^1–8 Recently, besides the conventional inorganic semiconductors (CdS and TiO₂), a number of all-inorganic and hybrid perovskites have been identified as visible light photocatalysts for photocatalytic H₂ evolution via splitting HX (X = I or Br) in aqueous solution.^9–11 Among them, methylammonium lead iodide (MAPbI₃), as a typical organic–inorganic hybrid perovskite, has been most widely studied for photocatalytic H₂ evolution due to its suitable band gap and excellent light absorption capability.¹² However, the rapid recombination of photogenerated electrons and holes in pristine MAPbI₃ significantly suppresses its photocatalytic performance. To overcome this limitation, two main strategies, decorating the surface with cocatalysts (MoS₂,¹³ CoP,¹⁴ Ni₃C,¹⁵ MoC,¹⁶etc.) and engineering the band gap and charge transfer kinetics by halide substitution (MAPbBr_3−xI_x),¹⁷ have been adopted to further improve the photocatalytic H₂ evolution performance of MAPbI₃. In particular, it has been demonstrated that compositing MAPbI₃ with conductive materials such as reduced graphene oxide (rGO),¹⁸ black phosphorus (BP),¹⁹ and Ti₃C₂T_x MXene²⁰ is also an effective route to develop high-performance MAPbI₃-based composite photocatalysts with enhanced charge separation efficiency and greatly improved photocatalytic performance for H₂ evolution compared with the pristine MAPbI₃.

Recently, owing to their high electrical conductivity and excellent chemical stability, conductive polymers such as polyaniline (PANI),²¹ polypyrrole (PPy),²² polythiophene (PTh),²³ poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS),²⁴ and their derivates have been extensively used as efficient electron or hole transfer media to boost the charge separation and thus enhance the photoelectrochemical conversion efficiency of perovskite solar cells. Of note, PANI is an abundant, low cost, and easily processable material that has also been employed as an efficient electron/hole acceptor and transporter to enhance the performance of a semiconductor photocatalyst by improving the charge separation.²⁵ In addition, the PANI chains after doping bear a large number of positively charged –NH₂⁺– groups that can effectively couple with hybrid perovskites through a strong ionic bond to form a strongly coupled composite.²¹ Moreover, the N atoms of amine groups on PANI chains bear abundant lone pair electrons, which are able to effectively capture H⁺ to facilitate the H₂ evolution reaction on perovskite photocatalysts.²⁶ Given the advantages of PANI, it could be envisioned that compositing conductive PANI with MAPbI₃ would result in an efficient composite photocatalyst with enhanced charge separation and photocatalytic H₂ evolution performance; however, there are still no reports on using PANI as a charge separation promoter to enhance the photocatalytic H₂ evolution performance of the MAPbI₃ microcrystals.

We report herein a 3D/1D heterostructured composite photocatalyst by coupling 3D MAPbI₃ microcrystals with 1D PANI NWs via an in situ doping-induced assembly method for efficient photocatalytic H₂ evolution from MAPbI₃-saturated HI solution under visible light irradiation. The strongly anchored conductive PANI NWs can greatly improve the charge separation of MAPbI₃. Consequently, the resulting MAPbI₃–PANI NW composite photocatalyst with an optimal mass ratio of MAPbI₃ to PANI exhibits a 29 times higher photocatalytic H₂ evolution rate than pristine MAPbI₃ microcrystals and relatively good stability.

The MAPbI₃ microcrystals were first synthesized according to our previously reported procedures (see experimental details in the ESI†) and characterized by X-ray diffraction (XRD), revealing that MAPbI₃ microcrystals are well-crystallized in the tetragonal phase without any other impurities (Fig. S1, ESI†). The HCl-doped PANI NWs were prepared through the facile interfacial polymerization method at room temperature (see experimental details in the ESI†). The dedoped PANI NWs were then obtained by reacting HCl-doped PANI NWs with NH₃·H₂O (1 M) followed by washing with water (Fig. S2, ESI†). Afterward, a series of MAPbI₃–PANI NW composite photocatalysts were prepared by directly mixing 100 mg of MAPbI₃ microcrystals with different amounts of dedoped PANI NWs in a MAPbI₃-saturated aqueous HI/H₃PO₂ (v/v = 4/1) solution and directly used for the photocatalytic H₂ evolution reaction without separation. Given that the MAPbI₃–PANI NW composite photocatalyst prepared with 100 mg of MAPbI₃ and 7 mg of PANI NWs exhibits the highest H₂ evolution activity (see the results and discussion below), the following discussion is mainly on this sample unless noted.

The XRD patterns (Fig. S1, ESI†) reveal that the resulting MAPbI₃–PANI NW composite photocatalyst has the same tetragonal structure and good crystallinity as pristine MAPbI₃ microcrystals, suggesting that the surface anchoring of PANI NWs can hardly change the bulk phase structure of the MAPbI₃. In addition, no characteristic diffraction peaks related to PANI NWs can be found in the XRD pattern of the composite probably due to the low content and low crystallinity of PANI NWs. Fig. 1 shows the scanning electron microscopy (SEM) images of the pristine MAPbI₃, PANI NWs, and MAPbI₃–PANI NW composite photocatalyst. The pristine MAPbI₃ possesses a cube-like morphology with a size of ∼10 μm and a relatively smooth surface (Fig. 1a), and the PANI NWs are short nanowires with diameters of 30–50 nm and lengths of 100–150 nm (Fig. 1b and S3, ESI†). For the MAPbI₃–PANI NW composite photocatalyst, the surface of the MAPbI₃ microcrystal becomes rougher (Fig. 1c) and a large number of PANI NWs are found to be firmly embedded on its surface (Fig. 1d). The corresponding energy dispersive X-ray (EDX) elemental maps (Fig. 1e) show a uniform distribution of I, Pd, and C on MAPbI₃ particles and the distribution of N mainly occurs at the corner of MAPbI₃ microcrystals, confirming the successful coupling of MAPbI₃ and PANI NWs. As a result, the specific surface area of MAPbI₃–PANI NW is enhanced to 3.5 m² g⁻¹, much higher than that of pristine MAPbI₃ (0.1 m² g⁻¹) due to the presence of high-surface-area PANI NWs (57.4 m² g⁻¹) (Fig. S4, ESI†). The transmission electron microscopy (TEM) image in Fig. 1f indicates that the PANI NWs are in close contact with MAPbI₃, and some PANI NWs are embedded inside the MAPbI₃ microcrystals (Fig. S5, ESI†). The high-resolution TEM (HRTEM) image (inset of Fig. 1f) further indicates that an intimate interface is formed between well-crystallized MAPbI₃ and amorphous PANI NWs, further revealing the formation of a strongly coupled MAPbI₃–PANI NW composite photocatalyst. The surface anchoring of PANI NWs on MAPbI₃ microcrystals was further studied by Fourier transform infrared spectroscopy (FTIR) (Fig. 1g), which reveals that all the characteristic peaks of PANI NWs and MAPbI₃ can be found in the FTIR spectrum of the MAPbI₃–PANI NW composite photocatalyst, further confirming the presence of PANI NWs in the composite.²⁷ The UV-vis diffuse reflectance spectra (Fig. 1h) indicate that the MAPbI₃–PANI NW composite photocatalyst shows greatly enhanced light absorption towards the near-infrared region compared to pristine MAPbI₃ due to the presence of PANI NWs. In addition, based on the corresponding Tauc plots (Fig. S6, ESI†), the band gap (E_g) of MAPbI₃–PANI NW (1.26 eV) is much smaller than that of pristine MAPbI₃ (1.49 eV) as a result of strong coupling interactions between MAPbI₃ and PANI NWs. These results reveal that coupling PANI NWs with MAPbI₃ would be effective in maximizing the light utilization efficiency and thus improving the photocatalytic H₂ evolution activity.^16,20

Fig. 1 SEM images of (a) a pristine MAPbI₃ microcrystal, (b) PANI NWs, and (c and d) the MAPbI₃–PANI NW composite photocatalyst. (e) EDX elemental maps (I, Pb, C and N) of the MAPbI₃–PANI NW composite photocatalyst. (f) Typical TEM and HRTEM (inset) images of the MAPbI₃–PANI NW composite photocatalyst. (g) FTIR spectra and (h) UV-vis diffuse reflectance spectra of the PANI NWs, pristine MAPbI₃ microcrystals, and MAPbI₃–PANI NW composite photocatalyst.

The successful combination of MAPbI₃ and PANI NWs was further illustrated by X-ray photoelectron spectroscopy (XPS). As shown in Fig. S7, ESI,† all of the C, N, I and Pb elements can be found in the XPS survey spectrum of the MAPbI₃–PANI NW composite photocatalyst, confirming the existence of PANI NWs. The high-resolution C 1s spectra in Fig. 2a indicate that the C=C peak of PANI NWs in the MAPbI₃–PANI NW composite photocatalyst presents a shift of 0.66 eV toward a lower binding energy as compared to that of pristine PANI NWs, and a more pronounced downshift (∼1.17 eV) is also found for the –NH₂⁺– in the N 1s spectra (Fig. 2b). These results clearly suggest the formation of chemical interactions between PANI NWs and MAPbI₃ microcrystals. Moreover, it should be noted that the –NH– peak of PANI NWs disappears in the composite, while the –NH⁺– peak is shifted to a higher binding energy by 0.43 eV.²⁷ This can be ascribed to the in situ doping of PANI NWs when mixing with MAPbI₃ in acidic HI/H₃PO₂ solution, which would effectively induce the assembly of MAPbI₃ and PANI NWs to form a strongly coupled composite photocatalyst. In addition, as shown in Fig. 2c and S8, ESI,† the Pb 4f and I 3d peaks for the MAPbI₃–PANI NW composite photocatalyst display shifts toward higher binding energies by 0.23 and 0.15 eV relative to pristine MAPbI₃, respectively. This further suggests that the PANI NWs are strongly anchored on the surface of MAPbI₃ microcrystals, leading to an efficient charge transfer between MAPbI₃ and PANI NWs, which would greatly improve the charge separation and thus enhance the photocatalytic performance of the MAPbI₃–PANI NW composite photocatalyst.¹⁶

Fig. 2 (a) C 1s and (b) N 1s XPS spectra of the PANI NWs, pristine MAPbI₃ and MAPbI₃–PANI NW composite photocatalyst. (c) Pb 4f XPS spectra of the pristine MAPbI₃ and MAPbI₃–PANI NW composite photocatalyst.

The photocatalytic H₂ evolution performance of the as-prepared MAPbI₃–PANI NW composite photocatalysts was evaluated in an MAPbI₃-saturated aqueous HI/H₃PO₂ mixed solution under visible light irradiation (see experimental details in the ESI†). Fig. 3a shows the photocatalytic H₂ evolution rates on MAPbI₃–PANI NW composite photocatalysts prepared from 100 mg of MAPbI₃ with different adding amounts of PANI NWs (2 to 10 mg), and those on pristine MAPbI₃ and PANI NWs are included for comparison. Clearly, the pristine MAPbI₃ shows a very low H₂ evolution rate (1.32 μmol h⁻¹). This is due to the fast charge recombination. On the other hand, the PANI NWs alone also show no activity toward H₂ evolution under the same measurement conditions. Impressively, when PANI NWs were anchored on MAPbI₃, the in situ formed MAPbI₃–PANI NW composite photocatalysts show greatly enhanced photocatalytic H₂ evolution activities compared to pristine MAPbI₃. This clearly suggests that the PANI NWs are effective as a charge separation promoter to boost the charge separation and thus the photocatalytic performance of MAPbI₃. The H₂ evolution rate is greatly enhanced with increasing adding amount of PANI NWs and reaches the highest value of 38.8 μmol h⁻¹ at 7 mg of PANI NWs, which is 29 times higher than that of pristine MAPbI₃. The total amount of H₂ evolved is 247.6 μmol and the turnover number (TON) is 3 in a 6 h reaction (Fig. S9, ESI†). The photocatalytic H₂ evolution activity of the MAPbI₃–PANI NW composite photocatalyst is also comparable to those of MAPbI₃-based photocatalysts (Table S1, ESI†). Additionally, the apparent quantum efficiency of H₂ evolution over this best photocatalyst is determined to be 0.3% at 420 nm (see details in the ESI†). Further increasing the adding amount of PANI NWs to 8 mg or more leads to a decreased performance since excessively anchored PANI NWs may interfere with the light absorption of MAPbI₃. In addition, it should be mentioned that when PANI NWs were replaced by equal amounts of the HCl-doped PANI NWs to prepare the MAPbI₃–PANI NW composite photocatalyst and other experimental conditions were maintained, the obtained H₂ evolution rate dramatically reduced to 4.8 μmol h⁻¹ (Fig. S10, ESI†). This can be explained by the fact that the HCl-doped PANI NWs are negatively charged (−10.9 mV) (Fig. S11, ESI†), and do not effectively couple with negatively charged MAPbI₃ (−17.4 mV) in MAPbI₃-saturated HI solution,¹⁵ thus leading to insufficient activity. In contrast, the dedoped PANI NWs are positively charged (+2.9 mV) and can readily couple with MAPbI₃via electrostatic interactions to form a MAPbI₃–PANI NW composite photocatalyst with strongly coupled interfaces, which are crucial to enhance the photocatalytic activity.

Fig. 3 (a) The rates of photocatalytic H₂ evolution on the pristine PANI NWs, pristine MAPbI₃, and as-prepared MAPbI₃–PANI NW composite photocatalysts. (b) Photocatalytic H₂ evolution stability of the MAPbI₃–PANI NW composite photocatalyst. Reaction conditions: MAPbI₃-saturated aqueous HI/H₃PO₂ mixed solution, 5 mL; light source, 10 W LED lamp, 370≤λ ≤ 780 nm. (c) PL spectra of the pristine MAPbI₃ and MAPbI₃–PANI NW composite photocatalyst. (d) Transient photocurrent response profiles of the pristine MAPbI₃ and MAPbI₃–PANI NW composite photocatalyst under visible light irradiation. (e) EIS Nyquist plots of the pristine MAPbI₃ and MAPbI₃–PANI NW composite photocatalyst. (f) Energy diagram and the proposed photocatalytic mechanism for the visible-light-driven H₂ evolution reaction on the MAPbI₃–PANI NW composite photocatalyst.

The photocatalytic H₂ evolution stability of the MAPbI₃–PANI NW composite photocatalyst was then tested by performing a continuous cycling reaction for a 30 h irradiation (with each cycle lasting 6 h). As shown in Fig. 3b, the H₂ evolution activity is quite stable for the first three cycles, and then gradually decreases during the last two cycles. After the 30 h reaction, the colour of the reaction solution turned light yellow (Fig. S12, ESI†), indicating the partial decomposition of MAPbI₃, as evidenced by XRD analysis (Fig. S13, ESI†). In addition, despite there being no obvious changes in the chemical composition and valence states of the composite (Fig. S14, ESI†), the SEM analysis (Fig. S15, ESI†) indicates that the MAPbI₃–PANI NW composite photocatalyst breaks into small pieces and the intimate contact between MAPbI₃ and PANI NWs is significantly damaged after the stability test. This leads to an unfavourable charge transfer from MAPbI₃ to PANI NWs, thereby resulting in inferior stability.

To clarify the role of PANI NWs in improving the photocatalytic performance of the MAPbI₃–PANI NW composite photocatalyst, the photoluminescence (PL) spectra of the samples were first measured. As presented in Fig. 3c, the pristine MAPbI₃ microcrystals exhibit an intense PL emission at 750 nm upon excitation at 500 nm originating from the serious electron–hole recombination, which significantly quenches after anchoring PANI NWs. This suggests that a more efficient electron–hole separation is achieved in the MAPbI₃–PANI NW composite photocatalyst.¹⁸ In addition, in comparison with the pristine MAPbI₃, the photocurrent response of the MAPbI₃–PANI NW composite photocatalyst is greatly enhanced (Fig. 3d). Mott–Schottky (MS) measurements were then performed to determine the conduction band (CB) and valence band (VB) levels of pristine MAPbI₃ and MAPbI₃–PANI NW, and the flat-band potentials (E_fb) can be used to estimate the CB potentials (E_CB). As shown in Fig. S16, ESI,† the E_CB of pristine MAPbI₃ and MAPbI₃–PANI NW was determined to be −3.97 and −3.96 eV vs. the vacuum energy level, respectively. Based on the above results and the bandgap energy obtained by UV-vis-DRS, the VB potential (E_VB) of pristine MAPbI₃ and MAPbI₃–PANI NW composite was calculated to be −5.46 and −5.20 eV vs. the vacuum energy level, respectively. Therefore, in combination with the reported E_VB for PANI (−5.1 eV vs. vacuum energy level),²¹ it can be inferred that the photogenerated holes produced on MAPbI₃ can migrate to PANI NWs to enable more efficient separation and transportation of the charges generated on MAPbI₃.²¹ These favorable effects can be attributed to the presence of PANI NWs in the composite, which can be expected to provide conductive pathways for fast charge transfer, which is further confirmed by electrochemical impedance spectra (EIS). As shown in Fig. 3e, the MAPbI₃–PANI NW composite photocatalyst exhibits a much smaller semicircle diameter and a much lower interfacial charge-transfer resistance (R_ct: 6.7 kΩ) compared to pristine MAPbI₃ (R_ct: 14.0 kΩ), indicating that the coupled PANI NWs offer fast conductive pathways to boost charge transfer and would thus be expected to improve the charge separation.^13,17

Based on the above experimental results, a photocatalytic H₂ evolution mechanism on the MAPbI₃–PANI NW composite photocatalyst is proposed, as illustrated in Fig. 3f. Upon visible light excitation, the photogenerated holes produced on MAPbI₃ quickly transfer to the PANI NWs across the intimate interface between the two, and an efficient separation of electrons and holes is achieved. Since the electrochemical measurements show that the PANI NWs can hardly catalyze the proton reduction reaction but are an efficient catalyst for the I⁻ oxidation reaction with a lower overpotential (Fig. S17 and S18, ESI†), the captured photogenerated holes by PANI NWs can be efficiently consumed for the I⁻ oxidation to I₃⁻. On the other hand, it has been verified that the E_CB of MAPbI₃ is located at ca. −3.96 eV versus the vacuum level, and thus the photogenerated electrons left on MAPbI₃ will have sufficient driving force to reduce protons to H₂.

In summary, we report a 3D/1D heterostructured composite photocatalyst by strongly coupling MAPbI₃ microcrystals with PANI NWs for photocatalytic H₂ evolution via splitting HI in MAPbI₃-saturated aqueous solution under visible light irradiation. The most active MAPbI₃–PANI NW exhibit a 29 times higher H₂ evolution rate than that of pristine MAPbI₃ and relatively good stability. The conductive PANI NWs provide fast charge transfer pathways that greatly improve the charge separation and enhance the photocatalytic H₂ evolution activity. This work confirms the effectiveness of using conductive polymers as charge transfer promoters to enhance the photocatalytic performance of MAPbI₃ and the new insights obtained can be extended to develop other all-inorganic and hybrid perovskite-based photocatalysts with enhanced performances.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No. 22162001), the Natural Science Foundation of Ningxia Province (Grant No. 2021AAC03194), the Graduate Student Innovation Project of North Minzu University (Grant No. YCX21138), the Foundation of Academic Top-notch Talent Support Program of North Minzu University (Grant No. 2019BGBZ08), the Leading Talents Program of Science and Technology Innovation in Ningxia Province (Grant No. 2020GKLRLX14), the Innovation and Entrepreneurship Projects for Returnees of Ningxia Province, and the Cooperative Scientific Research Project of Chunhui Plan of Ministry of Education of China (Grant No. 201900081).

References

1. D. D. Ren, W. N. Zhang, Y. N. Ding, R. C. Shen, Z. M. Jiang, X. Y. Lu and X. Li, Solar RRL, 2020, 4, 1900423.
2. R. C. Shen, K. L. He, A. P. Zhang, N. Li, Y. H. Ng, P. Zhang, J. Hu and X. Li, Appl. Catal. B Environ., 2021, 291, 120104.
3. H. T. Xu, R. Xiao, J. R. Huang, Y. Jiang, C. X. Zhao and X. F. Yang, Chin. J. Catal., 2021, 42, 107.
4. D. D. Ren, Z. Z. Liang, Y. H. Ng, P. Zhang, Q. J. Xiang and X. Li, Chem. Eng. J., 2020, 390, 124496.
5. C. Liu, Y. Feng, Z. T. Han, Y. Sun, X. Q. Wang, Q. F. Zhang and Z. G. Zou, Chin. J. Catal., 2021, 42, 164.
6. H. T. Xu, R. Xiao, J. R. Huang, Y. Jiang, C. X. Zhao and X. F. Yang, Chin. J. Catal., 2021, 42, 107.
7. F. F. Mei, Z. Li, K. Dai, J. F. Zhang and C. H. Liang, Chin. J. Catal., 2020, 41, 41.
8. J. R. Ran, J. Zhang, J. G. Yu, M. Jaroniecc and S. Z. Qiao, Chem. Soc. Rev., 2014, 43, 7787.
9. X. M. Wang, H. Wang, H. F. Zhang, W. Yu, X. L. Wang, Y. Zhao, X. Zong and C. Li, ACS Energy Lett, 2018, 3, 1159.
10. W. B. Li, S. X. Min, F. Wang, Z. G. Zhang and D. W. Gao, Chem. Commun., 2021, 57, 1121.
11. Z. H. Guan, Y. Q. Wu, P. Wang, Q. Q. Zhang, Z. Y. Wang, Z. K. Zheng, Y. Y. Liu, Y. Dai, M. H. Whangbo and B. B. Huang, Appl. Catal. B Environ., 2019, 245, 522.
12. S. Park, W. J. Chang, C. W. Lee, S. Park, H.-Y. Ahn and K. T. Nam, Nat. Energy, 2016, 2, 16185.
13. F. Wang, X. Y. Liu, Z. G. Zhang and S. X. Min, Chem. Commun., 2020, 56, 3281.
14. C. Cai, Y. Teng, J. H. Wu, J. Y. Li, H. Y. Chen, J. H. Chen and D. B. Kuang, Adv. Funct. Mater., 2020, 30, 2001478.
15. Z. J. Zhao, J. J. Wu, Y. Z. Zheng, N. Li, X. T. Li and X. Tao, ACS Catal., 2019, 9, 8144.
16. T. T. Zhang, J. F. Yu, J. Y. Huang, S. N. Lan, Y. B. Lou and J. X. Chen, Dalton Trans., 2021, 50, 10860.
17. Y. Q. Wu, P. Wang, Z. H. Guan, J. X. Liu, Z. Y. Wang, Z. K. Zheng, S. Y. Jin, Y. Dai, M. H. Whangbo and B. B. Huang, ACS Catal., 2018, 8, 10349.
18. Y. Q. Wu, P. Wang, X. L. Zhu, Q. Q. Zhang, Z. Y. Wang, Y. Y. Liu, G. Z. Zou, Y. Dai, M. H. Whangbo and B. B. Huang, Adv. Mater., 2018, 30, 1704342.
19. R. Li, X. T. Li, J. J. Wu, X. D. Lv, Y. Z. Zheng, Z. J. Zhao, X. Q. Ding, X. Tao and J. F. Chen, Appl. Catal., B, 2019, 259, 118075.
20. W. B. Li, F. Wang, Z. G. Zhang and S. X. Min, Chem. Commun., 2021, 57, 7774.
21. J. W. Wei, F. R. Huang, S. N. Wang, L. Y. Zhou, Y. L. Xin, P. Jin, Z. Cai, Z. D. Yin, Q. Pang and J. Z. Zhang, Mater. Res. Bull., 2018, 106, 35.
22. H. Zarenezhada, T. Balkana, N. Solati, M. Halali, M. Askarie and S. Kaya, Sol. Energy, 2020, 207, 1300.
23. H. C. Hsieh, C. Y. Hsiow, Y. A. Suc, Y. C. Liu, W. Chen, W. Y. Chiu, Y. C. Shih, K. F. Lin and L. Wang, J. Power Sources, 2019, 426, 55.
24. H. Wang, X. M. Wang, R. T. Chen, H. F. Zhang, X. L. Wang, J. H. Wang, J. Zhang, L. C. Mu, K. F. Wu, F. T. Fan, X. Zong and C. Li, ACS Energy Lett., 2019, 4, 40.
25. Z. G. Zhang, G. R. Xie, F. Wang and S. X. Min, Chem. Commun., 2021, 57, 663.
26. J. X. Feng, S. Y. Tong, Y. X. Tong and G. R. Li, J. Am. Chem. Soc., 2018, 140, 5118.
27. T. Glaser, C. Müller, M. Sendner, C. Krekeler, O. E. Semonin, T. D. Hull, O. Yaffe, J. S. Owen, W. Kowalsky, A. Pucci and R. Lovrinčić, J. Phys. Chem. Lett., 2015, 6, 2913.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and addition data. See DOI: 10.1039/d1se01680f

---