A sustainable solvent system for processing CsPbBr₃ films for solar cells via an anomalous sequential deposition route

Abstract

Toxic solvents used in the fabrication of perovskite films are a non-ignorable obstacle for the commercialization of perovskite solar cells (PSCs). CsPbBr₃ based solar cells have attracted increasing attention due to their high stability in ambient air. Usually, a traditional sequential deposition route is employed to fabricate CsPbBr₃ films, in which toxic DMF and methanol are separately used to dissolve PbBr₂ and CsBr. In order to eliminate the solvent toxicity during the fabrication of CsPbBr₃ films, we develop an anomalous sequential deposition route to prepare CsPbBr₃ films. In this new route, a green solvent system composed of water with the addition of appropriate ethanol and ethylene glycol (EG) is developed to prepare the CsBr solution, and a green solvent, triethyl phosphate (TEP), is used to prepare the PbBr₂ solution. Different from those obtained by the traditional sequential deposition route, CsBr films are firstly prepared from their solution, then reacted with PbBr₂ through a dynamic spin-coating process, and subsequently transformed to CsPbBr₃ films via annealing treatment. As a result, solar cells based on the device configuration of FTO/TiO₂/CsPbBr₃/carbon exhibit a power conversion efficiency of 6.86%, which is comparable to that of the PSCs prepared through the traditional sequential deposition route using toxic solvents. This work provides an environment-friendly route to prepare CsPbBr₃ perovskite solar cells without efficiency loss.

---

1. Introduction

Perovskite solar cells (PSCs) have attracted much attention owing to their outstanding performance and easy fabrication process. The power conversion efficiency (PCE) of PSCs has been boosted from the initial 3.8% to 25.5% in a decade.^1,2 In solution-processed methods, perovskite precursors were dissolved in organic solvents or their mixtures, and then deposited onto conductive substrates. N,N-Dimethylformamide (DMF), dimethyl sulfoxide (DMSO), N-methyl-2-pyrrolidone (NMP) and N,N-dimethylacetamide (DMAc) are the most commonly used solvents during the fabrication of perovskite films.³ Meanwhile, toluene⁴ and chlorobenzene (CB)⁵ are generally used as anti-solvents to control the crystallization of perovskites, so as to improve the quality of the perovskite films. However, these solvents are demonstrated to be harmful to human health because of their toxicity. Recently, researchers have begun to address the issue of highly toxic anti-solvents by replacing the highly toxic anti-solvents with low toxicity/non-toxic solvents. Various non-toxic solvents, including anisole,^6–9 ethyl acetate,¹⁰ tetraethyl orthosilicate,¹¹ and methyl benzoate,¹² have been chosen to replace the traditional toxic anti-solvents. However, the toxic solvents used for dissolving the perovskite precursors have received limited attention owing to the low solubility of the perovskite precursors in the commonly used green solvents.

CsPbBr₃-based PSCs have attracted increasing attention owing to their high stability in air. The quality of CsPbBr₃ films, including the morphology, crystallization, and grain size, is key for achieving highly efficient PSCs. Generally, CsPbBr₃ films are fabricated by traditional^13,14 or modified^15–17 sequential deposition routes, where PbBr₂ films were firstly deposited onto substrates from PbBr₂/DMF solution in the first step. Then, the PbBr₂ films were exposed to CsBr/methanol solution in the second step to form CsPbBr₃ films. In the traditional sequential deposition route, toxic DMF and methanol were separately used to dissolve PbBr₂ and CsBr, which are potentially hazardous to human health and the environment. In order to minimize the consumption of toxic solvents, we used water to replace methanol to prepare high quality CsPbBr₃ films in the second step.¹⁸ However, we have no choice but to use toxic DMF for preparing PbBr₂ films in the first step owing to the low solubility of PbBr₂ in the commonly used green solvents. Chronic DMF exposure can cause damage to various organs, especially to the liver.¹⁹ Statistical data show that the workers working in the synthetic textile industry show more serious liver function injury than the general population due to their long-term exposure to the DMF environment.²⁰ Therefore, it is urgent to develop a green solvent system to prepare CsPbBr₃ films, which guarantees operators' health.

In order to fabricate CsPbBr₃ films from all green solvents, we propose an anomalous sequential deposition route in this work. In this novel method, fully covered CsBr films are firstly deposited onto substrates by spin-coating from CsBr solution, which employs the addition of water with some ethanol and ethylene glycol (EG) as green solvents. Then, a few drops of PbBr₂ solution, which uses triethyl phosphate (TEP) as the green solvent, are spin-coated onto the CsBr films. Finally, the CsPbBr₃ films are obtained by annealing. As a result, a PCE of 6.86% is obtained based on a typical device configuration of FTO/TiO₂/CsPbBr₃/carbon, which is comparable to those prepared by the traditional sequential deposition method using toxic solvents.

2. Experimental

The fabrication and characterization of devices are conducted under open air conditions. Fluorine-doped tin oxide (FTO) glass was cleaned with deionized water, acetone 2-propanol, ethanol, and deionized water in turn under ultrasonic conditions. The glasses were then exposed to a UV ozone cleaner for 15 min at 100 °C before use. Then, an ethanol solution of titanium isopropoxide was spin-coated onto the FTO substrates at a speed of 2000 rpm for 30 s, followed by annealing at 500 °C for 30 min to form a compact TiO₂ layer. After cooling to room temperature, diluted TiO₂ paste (Dyesol-18NRT, Dyesol) in ethanol with a weight ratio of 2 : 10 was spin-coated onto the compact TiO₂ layer at a speed of 3000 rpm for 30 s and then annealed at 500 °C for 30 min to prepare a mesoporous TiO₂ layer. Subsequently, the FTO/TiO₂ substrates were dipped into an aqueous solution of 0.04 M TiCl₄ for 30 min at 75 °C, and annealed at 450 °C for 30 min.

For comparison, CsPbBr₃ films were separately prepared by traditional and anomalous sequential deposition routes. In the traditional sequential deposition route (see Fig. 1a), PbBr₂/DMF solution (1.0 M) was spin-coated onto the mesoporous TiO₂ layer at 2000 rpm for 30 s, and then annealed at 90 °C for 30 min to form PbBr₂ films. Then, the PbBr₂ films were dipped in CsBr/methanol (15 mg mL⁻¹) at 50 °C for the optimized time. Subsequently, the Cs–Pb–Br films were annealed at 250 °C for 5 min. In the anomalous sequential deposition route (see Fig. 1b), CsBr films were first prepared by spin-coating from CsBr/H₂O solution (∼1.0 M) by adding appropriate ethanol and ethylene glycol. Then, 60 μL of PbBr₂/TEP (∼60 mg mL⁻¹) solution (see Fig. S1†) was added dropwise onto the CsBr films at 2000 rpm in the last 10 s. Finally, the films were annealed at 250 °C for 5 min. All of the Cs–Pb–Br films were washed with 2-propanol before characterization. The molecular structures of solvents used in the fabrication of CsPbBr₃ films through different routes are listed in Fig. S2.† A carbon black electrode was deposited onto the CsPbBr₃ films by using screen printing to form a complete solar cell. The active area of the solar cells was fixed to 0.04 cm² during the photovoltaic performance measurement.

Fig. 1 Schematic illustration of the fabrication process of Cs–Pb–Br films through different routes: (a) traditional route using toxic solvents and (b) anomalous route using green solvents.

Characterization

The samples were separately characterized by X-ray diffraction (XRD, PANalytical X'Pert3 Powder), field-emission scanning electron microscopy (SEM, Nova NanoSEM 430), and UV-VIS spectrophotometry (UV-2550). The current density–voltage (J–V) curves were obtained under AM 1.5G simulated solar illumination, and recorded using a Keithley 2400 source meter. Incident photon-to-current conversion efficiency (IPCE) spectra were recorded using an integrated system from Enlitech (QE-R3011). Steady and time resolved photoluminescence (PL) spectra were obtained using a photoluminescence spectrometer instrument (Edinburgh Instruments, LP 980). Impedance spectroscopy (IS) spectra were recorded using an electrochemical workstation (CHI660D) under dark conditions.

3. Results and discussion

In light of the high solubility of CsBr (∼5.8 M) in water at room temperature,²¹ we firstly prepared CsBr films using CsBr/H₂O (∼1.0 M) in the first step. Fig. 2a shows a SEM image of the CsBr film directly prepared from the CsBr/H₂O solution, showing a discontinuous film composed of some isolated grains. The discontinuous CsBr film might be ascribed to the poor wettability, which is confirmed from a large contact angle (∼30°) of water on substrates (see Fig. S3a†). In order to improve the wettability of the solution, we employ a green mixed solvent system composed of H₂O and ethanol (volume ratio of v/v = 1 : 1) to dissolve the CsBr (∼1.0 M). The contact angle almost decreases to zero (see Fig. S3b†). As shown in Fig. 2b, the CsBr transforms from isolated grains to a ribbon-like network. However, there still are large areas of voids in the CsBr films.

Fig. 2 SEM images of the CsBr films fabricated through different solvent systems in the first step: (a) H₂O, (b) H₂O + ethanol, and (c) H₂O + ethanol + EG. (d) Enlarged XRD patterns of the CsBr films.

In order to obtain continuous CsBr films, we introduce 10% EG into a mixture of H₂O and ethanol (v/v = 1 : 1) to prepare a CsBr solution. Surprisingly, it forms a desired continuous and smooth CsBr film with a full coverage surface (see Fig. 2c). The low magnification SEM images of CsBr films are also given in Fig. S4,† reconfirming the characteristic morphology on a large scale as discussed above. Fig. 2d shows the XRD curves of the CsBr films fabricated from different solutions. There are characteristic diffraction peaks of CsBr centered at 20.6°, 29.3° and 42.0° in all of the XRD curves (see Fig. S5†), which are consistent with a previous report.²² This implies that the films prepared from the CsBr solutions are composed of mainly CsBr crystals.

Fig. 3a–c show a photograph of the Cs–Pb–Br films fabricated from different solvent systems. The Cs–Pb–Br films fabricated from H₂O and a mixture of (H₂O + ethanol) mainly are light white decorated with some yellow dots (see Fig. 3a and b). However, the Cs–Pb–Br film fabricated from the mixture solvents of (H₂O + ethanol + EG) shows a uniform yellow morphology (see Fig. 3c). The corresponding SEM images of the three Cs–Pb–Br films are shown in Fig. 3d–f. The Cs–Pb–Br films fabricated from H₂O and ethanol solution without adding EG consist of many isolated grains (see Fig. 3d and e). However, the Cs–Pb–Br films fabricated from solvent systems of (H₂O + ethanol + EG) show a continuous morphology with full coverage (see Fig. 3f). The XRD curves of the Cs–Pb–Br films also are given in Fig. S6.† They show that only the films fabricated from CsBr/(H₂O + ethanol + EG) afford the desired pure CsPbBr₃ for application in solar cells. These results demonstrate that it is important to prepare continuous CsBr films in the first step from a suitable solvent system, which directly affects the morphology and phase of the resultant Cs–Pb–Br films. Hereafter, we choose the mixed solvents of H₂O + ethanol + EG to prepare CsBr and CsPbBr₃ films so as to fabricate perovskite solar cells.

Fig. 3 Photographs (top) and SEM images (bottom) and of the Cs–Pb–Br films fabricated through different solvent systems. (a) and (d) are from water; (b) and (e) are from H₂O + ethanol; (c) and (f) are from H₂O + ethanol + EG.

In order to compare the performance of the solar cells, the CsPbBr₃ layers are separately prepared through the traditional sequential deposition route using the toxic DMF solvent (see Fig. 1a) and the anomalous sequential deposition route using green solvents (see Fig. 1b). Hereafter, we lable the samples according to their fabrication route (Traditional or Anomalous) and solvent (Toxic solvents or Green solvents) for the following discussion. For the Cs–Pb–Br films fabricated through the traditional sequential deposition route, they were prepared using the optimized dipping time (9 min) based on the best morphology (see Fig. S7†) and pure phase of CsPbBr₃ (see Fig. S8†). Fig. 4a shows the XRD curves of the Cs–Pb–Br film fabricated through different deposition routes with the optimized fabrication conditions. There are evident characteristic diffraction peaks at 15.2°, 21.6° and 30.6° in the two XRD curves, corresponding to the (100), (110) and (200) planes of CsPbBr₃.¹⁶Fig. 4b shows the relationship between (Ahv)² and photoenergy (hv) of the CsPbBr₃ perovskite films. The bandgap of the CsPbBr₃ films is fitted to be ∼2.3 eV, which is consistent with the previous work.¹⁷Fig. 4c shows a high magnification SEM image of the CsPbBr₃ films fabricated through the traditional sequential deposition route. There are some voids in the films, which might lead to direct contact between TiO₂ and the carbon electrode. Fig. 4e shows such a short contact between the carbon electrode and the TiO₂ layer. Fig. 4d shows a top-view SEM image of the CsPbBr₃ film fabricated through the anomalous sequential deposition route. It shows a full coverage morphology, which is also further confirmed from the cross-sectional SEM image of a complete PSC (see Fig. 4f). The fully covered CsPbBr₃ films in the solar cells can prevent the direct contact between carbon and TiO₂, which is strongly desired for fabricating highly efficient PSCs. It is noted that the thickness of the CsPbBr₃ films is about 260 nm. The thickness of the CsPbBr₃ films is mainly limited by the highest solubility of PbBr₂ in TEP (∼60 mg mL⁻¹). It will produce an undesired phase, including Cs₄PbBr₆ and residual CsBr in the Cs–Pb–Br films if a higher concentration of CsBr solution (∼1.5 M) is used in the first step (see Fig. S9†).

Fig. 4 (a) XRD curves, (b) The bandgap of CsPbBr₃ films fitted from UV-vis absorption spectra. The surface of CsPbBr₃ films and the cross-sectional SEM images of whole solar cells; (c) and (e) are fabricated through the traditional route using toxic solvents. (d) and (f) are fabricated through the anomalous route using green solvents.

Fig. 5a shows the light J–V curves of the best PSCs with a typical device configuration of FTO/TiO₂/CsPbBr₃/carbon (see Fig. 4e and f), where the CsPbBr₃ films are fabricated through two different deposition routes. The PSCs prepared through the anomalous sequential deposition route show a power conversion efficiency (PCE) of 6.86%, a short current density (J_sc) of 7.04 mA cm⁻², an open-circuit voltage (V_oc) of 1.30 V, and a fill factor (FF) of 0.750. The IPCE spectra of the solar cells are shown in Fig. S10.† It shows that they exhibit a quantum yield in the range from 300 nm to 539 nm, which is consistent with a bandgap of 2.3 eV (see Fig. 4b). The integrated J_sc from the IPCE spectrum is 5.37 mA cm⁻², which is lower than the value obtained from the reverse scan. This phenomenon is widely observed in the research of CsPbBr₃ based solar cells from several independent groups.^23–27 The discrepancy may be ascribed to the loss of ultraviolet spectra during IPCE characterization.^25,27 However, the best PSC prepared through the traditional sequential deposition route exhibits an inferior PCE of 6.15%, a J_sc of 6.53 mA cm⁻², a V_oc of 1.21 V and a FF of 0.778. Fig. 5b shows the stabilized outputs of the best PSCs at the maximum power points. The best PSC fabricated through the anomalous sequential deposition route has a stabilized current density of 5.83 mA cm⁻², delivering a stabilized PCE of 6.35%. However, the stabilized output current density and PCE are 5.02 mA cm⁻² and 5.36%, respectively, for the best PSC fabricated through the traditional sequential deposition route. It is noted that both the stabilized outputs are slightly lower than the value obtained from reverse scans (see Fig. 5a). This phenomenon is widely reported in the research of PSCs with hysteresis behaviors^28,29 (see Fig. S11†), which may come from the interface between FTO and the compact TiO₂ layer.³⁰ The detailed comparison of the photovoltaic parameters of the best PSCs between this work and previous reports with a typical device configuration of FTO/TiO₂/CsPbBr₃/carbon without any modification is listed in Table S1.† Meanwhile, the solvents used to prepare CsPbBr₃ films in these works are also listed in Table S1.† It clearly shows that the performance of the PSCs in this work is comparable to those fabricated using a toxic solvent system used by independent research groups. This strongly demonstrates that the anomalous sequential deposition route in this work is an effective and environmentally friendly approach to prepare CsPbBr₃-based solar cells without efficiency loss.

Fig. 5 Photovoltaic performance of PSCs fabricated through different routes. (a) Light J–V curves and (b) stabilized output at the maximum power of the best PSCs. (b) Box chart of photovoltaic performance, (c) PCE, (d) V_oc, (e) FF, and (f) J_sc.

Fig. 5c–f show the statistical results of the photovoltaic parameters based on a series of PSCs fabricated through different deposition routes. The superior PCE of PSCs fabricated through the anomalous sequential deposition route is mainly attributed to the significantly enhanced V_oc (see Fig. 5d). In order to explore the mechanisms of the V_oc improvement of PSCs, PL spectra of CsPbBr₃ films on glasses are recorded. As shown in Fig. 6a, both the CsPbBr₃ films have an emission peak at ∼527 nm, corresponding to a bandgap of ∼2.35 eV, which is consistent with the results obtained from the UV-Vis spectra (see Fig. 4b). However, the CsPbBr₃ films prepared through the anomalous route using green solvents exhibit higher PL intensity than that prepared through the traditional route using toxic solvents. The higher PL intensity indicates the low trap density of the CsPbBr₃ films prepared through the anomalous route using green solvents.¹⁵Fig. 6b shows the time-resolved PL spectra of CsPbBr₃ films prepared through different routes. All the curves can be well fitted by the exponential equation, , where τ₁ and τ₂ are the fast and slow decay times, and A₁ and A₂ are the relative amplitudes, respectively.³¹ The average life time (τ_ave) is calculated based on the equation, τ_ave = (A₁τ²₁ + A₂τ₂²)/(A₁τ₁ + A₂τ₂).³¹ The CsPbBr₃ films prepared through the anomalous route using green solvents exhibit an average lifetime of 3.36 ns, which is longer than the value (2.04 ns) obtained for the CsPbBr₃ films prepared through the traditional route using toxic solvents (see Table S2†). The longer lifetime reconfirms the low trap density of the CsPbBr₃ films prepared through the anomalous route using green solvents. Fig. 6c shows the impedance spectroscopy (IS) spectra of the best PSCs prepared through different routes. All PSCs exhibit an arc in the low frequency region, which reflects the recombination rate in PSCs. The larger the arc in the IS spectra, the lower the recombination rate in PSCs.³² As shown in Fig. 6c, the PSC prepared through the anomalous route using green solvents exhibits a larger arc than that prepared through the traditional route using toxic solvents, indicating a lower recombination rate in the solar cells. Therefore, the low recombination rate and fully covered CsPbBr₃ films contribute to the enhancement of V_oc in the PSCs prepared through the anomalous route using green solvents in comparison with the ones prepared through the traditional route using toxic solvents.

Fig. 6 (a) Steady and (b) time-resolved PL spectra of CsPbBr₃ films prepared through different routes. (c) Impedance spectroscopy spectra of the PSCs under a bias voltage at 1.0 V under dark conditions.

The stability of PSCs is a key factor for the industrialization of PSCs. We also characterize the stability of PSCs in air without any encapsulation. Fig. 7 shows the evolution of the normalized parameters of PSCs with the increase of the storage time in air (with a relative humidity of ∼40% and a temperature of ∼30 °C). This shows that the PCE still exhibits a comparable value to the initial value when the device was stored in air for 14 days, showing the excellent stability of CsPbBr₃ based PSCs. The excellent stability is mainly attributed to the high stability of CsPbBr₃^13,33and all inorganic functional layers in the solar cells.

Fig. 7 The evolution of normalized parameters of the PSCs under air conditions without encapsulation. (a) PCE; (b) V_oc; (c) J_sc; (d) FF.

4. Conclusions

Full coverage CsPbBr₃ films are prepared through an anomalous sequential deposition route, in which green solvents are used to prepare CsBr and PbBr₂ solutions. Different from those obtained by the traditional sequential deposition route, CsBr films are firstly deposited onto substrates, and then reacted with PbBr₂ to form full coverage CsPbBr₃ films through a dynamic spin-coating process. A power conversion efficiency of 6.86% is achieved for the solar cells with an unmodified configuration of FTO/TiO₂/CsPbBr₃/carbon, which is comparable to that of the solar cells prepared through the traditional sequential deposition route using toxic solvents. This work not only provides a new route to prepare CsPbBr₃ films but also addresses the solvent toxicity for processing CsPbBr₃ films. It is an environment-friendly route, which has great potential in fabricating CsPbBr₃-based devices, such as resistive memory, light-emitting diodes, photodetectors, and so on.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the Innovation Projects of Department of Education of Guangdong Province (2019KQNCX155), the Science Foundation for High-level Talents of Wuyi University (2019AL001), the National Natural Science Foundation of China (51972188), the Featured Innovation Projects of Colleges and Universities in Guangdong Province (2018KTSCX232), the Innovative Leading Talents of Jiangmen [Jaingren (2019) 7], the Science and Technology Projects of Jiangmen (No. (2017) 307, (2017) 149, (2018) 352), the Cooperative education platform of Guangdong Province (No. (2016) 31) and the Key Laboratory of Optoelectronic materials and Applications in Guangdong Higher Education (No: 2017KSYS011).

References

1. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
2. National Renewable Energy Laboratory Best Research-cell efficiencies. http://www.nrel.gov/pv/assets/images/efficiency-chart.png.
3. D. H. Kim, J. B. Whitaker, Z. Li, M. F. A. M. van Hest and K. Zhu, Joule, 2018, 2, 1437–1451.
4. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. Ryu and S. Seok, Nat. Mater., 2014, 13, 897–903.
5. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y. B. Cheng and L. Spiccia, Angew. Chem., Int. Ed., 2014, 53, 9898–9903.
6. P. Zhao, B. J. Kim, X. Ren, D. G. Lee, G. J. Bang, J. B. Jeon, W. B. Kim and H. S. Jung, Adv. Mater., 2018, 30, 1802763.
7. M. Zhang, Z. Wang, B. Zhou, X. Jia, Q. Ma, N. Yuan, X. Zheng, J. Ding and W. H. Zhang, Sol. RRL, 2018, 2, 1700213.
8. M. Yavari, M. Mazloum-Ardakani, S. Gholipour, M. M. Tavakoli, S. Turren-Cruz, N. Taghavinia, M. Grätzel, A. Hagfeldt and M. Saliba, Adv. Energy Mater., 2018, 8, 1800177.
9. D. Xin, Z. Wang, M. Zhang, X. Zheng, Y. Qin, J. Zhu and W. H. Zhang, ACS Sustainable Chem. Eng., 2019, 7, 4343–4350.
10. T. Bu, L. Wu, X. Liu, X. Yang, P. Zhou, X. Yu, T. Qin, J. Shi, S. Wang, S. Li, Z. Ku, Y. Peng, F. Huang, Q. Meng, Y. B. Cheng and J. Zhong, Adv. Energy Mater., 2017, 7, 1700576.
11. M. Wang, Q. Fu, L. Yan, P. Guo, L. Zhou, G. Wang, Z. Zheng and W. Luo, ACS Appl. Mater. Interfaces, 2019, 11, 3909–3916.
12. Y. Yun, F. Wang, H. Huang, Y. Fang, S. Liu, W. Huang, Z. Cheng, Y. Liu, Y. Cao, M. Gao, L. Zhu, L. Wang, T. Qin and W. Huang, Adv. Mater., 2020, 30, 1907123.
13. M. Kulbak, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2015, 6, 2452–2456.
14. X. Chang, W. Li, L. Zhu, H. Liu, H. Geng, S. Xiang, J. Liu and H. Chen, ACS Appl. Mater. Interfaces, 2016, 8, 33649–33655.
15. P. Teng, X. Han, J. Li, Y. Xu, L. Kang, Y. Wang, Y. Yang and T. Yu, ACS Appl. Mater. Interfaces, 2018, 10, 9541–9546.
16. X. Liu, X. Tan, Z. Liu, H. Ye, B. Sun, T. Shi, Z. Tang and G. Liao, Nano Energy, 2019, 56, 184–195.
17. J. Duan, Y. Zhao, B. He and Q. Tang, Angew. Chem., Int. Ed., 2018, 57, 3787–3791.
18. X. Cao, G. Zhang, L. Jiang, Y. Cai, Y. Gao, W. Yang, X. He, Q. Zeng, G. Xing, Y. Jia and J. Wei, ACS Appl. Mater. Interfaces, 2020, 12, 5925–5931.
19. M. Li and T. Zeng, Chem.-Biol. Interact., 2019, 298, 129–136.
20. R. Wrbitzky, Int. Arch. Occup. Environ. Health, 1999, 72, 19–25.
21. L. Wu, H. Hu, Y. Xu, S. Jiang, M. Chen, Q. Zhong, D. Yang, Q. Liu, Y. Zhao, B. Sun, Q. Zhang and Y. Yin, Nano Lett., 2017, 17, 5799–5804.
22. V. M. Brendel, S. V. Garnov, T. F. Yagafarov, L. D. Iskhakova and R. P. Ermakov, Quantum Electron., 2014, 44, 841–844.
23. D. Huang, P. Xie, Z. Pan, H. Rao and X. Zhong, J. Mater. Chem. A, 2019, 7, 22420–22428.
24. G. Tong, T. Chen, H. Li, L. Qiu, Z. Liu, Y. Dang, W. Song, L. K. Ono, Y. Jiang and Y. Qi, Nano Energy, 2019, 65, 104015.
25. G. Tong, T. Chen, H. Li, W. Song, Y. Chang, J. Liu, L. Yu, J. Xu, Y. Qi and Y. Jiang, Sol. RRL, 2019, 3, 1900030.
26. J. Lei, F. Gao, H. Wang, J. Li, J. Jiang, X. Wu, R. Gao, Z. Yang and S. F. Liu, Sol. Energy Mater. Sol. Cells, 2018, 187, 1–8.
27. H. Li, G. Tong, T. Chen, H. Zhu, G. Li, Y. Chang, L. Wang and Y. Jiang, J. Mater. Chem. A, 2018, 6, 14255–14261.
28. N. K. Noel, A. Abate, S. D. Stranks, E. S. Parrott, V. M. Burlakov, A. Goriely and H. J. Snaith, ACS Nano, 2014, 8, 9815–9821.
29. Y. Zhang, M. Liu, G. E. Eperon, T. C. Leijtens, D. McMeekin, M. Saliba, W. Zhang, M. Bastiani, A. Petrozza, L. M. Herz, M. B. Johnston, H. Lin and H. J. Snaith, Mater. Horiz., 2015, 2, 315–322.
30. A. K. Jena, H. W. Chen, A. Kogo, Y. Sanehira, M. Ikegami and T. Miyasaka, ACS Appl. Mater. Interfaces, 2015, 7, 9817–9823.
31. J. Chen, X. Zhao, S. G. Kim and N. G. Park, Adv. Mater., 2019, 31, 1902902.
32. X. Dong, D. Chen, J. Zhou, Y. Z. Zheng and X. Tao, Nanoscale, 2018, 10, 7218–7227.
33. M. Kulbak, S. Gupta, N. Kedem, I. Levine, T. Bendikov, G. Hodes and D. Cahen, J. Phys. Chem. Lett., 2016, 7, 167–172.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0gc02892d

---