Organic hole-transporting materials for 9.32%-efficiency and stable CsPbBr₃ perovskite solar cells

Abstract

Cost-effective and stable CsPbBr₃-based inorganic perovskite solar cells (PSCs) are regarded as promising candidates for next-generation photovoltaics. However, the large interfacial energy differences at the CsPbBr₃/hole-transporting layer lead to serious charge recombination and poor charge extraction kinetics. Herein, we prepare a series of hole-transporting materials (HTMs) to improve hole extraction and to reduce electron–hole recombination at the CsPbBr₃/HTM interface. In comparison with the power conversion efficiency (PCE) of 6.10% for an HTM-free device, the CsPbBr₃ PSCs with polymeric HTMs such as polythiophene, polypyrrole and polyaniline yield efficiencies of 8.36%, 8.32% and 7.69%, respectively. Similarly, the inorganic PSC with organic small molecule BT-BTH achieves a PCE as high as 9.32% due to the improved hole conductivity. Moreover, the unencapsulated PSC with BT-BTH maintains 94% of its initial efficiency in 70% relative humidity over 80 days.

---

Introduction

Organic–inorganic hybrid perovskite solar cells (PSCs) have attracted considerable interest due to their low-cost fabrication and high photovoltaic performances.^1–3 Grätzel et al. designed the first all-solid state PSCs using CH₃NH₃PbI₃ quantum dots as light harvesting materials, yielding a power conversion efficiency (PCE) of 9.7% under one sun illumination.⁴ To date, the maximized certified PCE has been 23.3% arising from optimization of the perovskites, processing technology and device structure.⁵ Although the PCE of hybrid PSCs has exhibited rapid growth over the past few years, the intrinsic instability of organic–inorganic perovskites and iodide diffusion are still challenging for their commercialization.^6–10

Recently, all-inorganic CsPbBr₃ perovskite has been regarded as a promising light-harvester because of its high environmental tolerances and stabilized crystal structure.^11,40 However, the first CsPbBr₃ based PSC with a classical configuration of FTO/TiO₂/CsPbBr₃/carbon has a PCE of only 6.7%,¹² which is much lower than those of the state-of-the-art hybrid PSCs. There are two main reasons for the low efficiency of CsPbBr₃ PSCs. On the one hand, the bandgap as large as 2.3 eV for CsPbBr₃ results in a narrow absorption wavelength below 550 nm. On the other hand, the high energy difference at the CsPbBr₃/carbon interface (0.6 eV) causes serious electron–hole recombination and poor charge extraction.¹³ Although the substitution of Br⁻ with I⁻ ions to form mixed CsPbI_3−xBr_x (x = 0, 1, 2) halides provides a way of reducing perovskite bandgaps and energy differences at interfaces, the environmental tolerances especially in high humidity are also brought down. A solution to this impasse is to set an intermediate energy by introducing a hole-transporting layer (HTL) or coating quantum dots¹⁴ between CsPbBr₃ and the carbon electrode to reduce interface recombination loss and to promote charge extraction. The low coverage of quantum dots on CsPbBr₃ film does not maximize the hole extraction ability, therefore the addition of a HTM is more feasible in charge-transporting theory and solar cell structure.

Hole-transporting materials (HTMs) are good conductors that enable photo-induced holes to be efficiently extracted from the perovskite layer to the back electrodes. This requires HTMs to possess excellent charge-carrier mobility and suitable energy levels to match well with the perovskite halide and back electrode. Small-molecule HTMs made by molecular engineering provide advantages of monodispersity, high purity and good reproducibility in synthesis, and tunable energy levels, charge mobility and hole conductivity in physicochemical performances.

The state-of-the-art HTMs such as 2,2′,7,7′-tetrakis-(N,N-di-p-methoxyphenylamine)-9,9′-spiro biuorene (spiro-MeOTAD) and poly-triarylamine (PTAA) are still economic burdens for their commercial applications in PSCs.^15–22 Such HTMs generally need to be doped with lithium ions because of their poor conductivity, however, the strong hygroscopicity of dopant lithium ions simultaneously increases the instability of these HTMs. Alternatively, other cost-effective p-type conducting polymers such as poly 3,4-ethylenedioxythiophene (PEDOT), polypyrrole (PPy) and polyaniline (PANi) can be employed as electron barrier layers without dopant due to their high hole mobility from the conjugated structure. Nevertheless, the long molecular chains and inhomogeneity of polymers are detrimental to film forming properties. Therefore, it is imperative to develop undoped organic molecular materials.

In this work, we present the synthesis of an organic small molecule BT-BTH [2-(3,5-bis(5-(5-hexylthiophen-2-yl)thiophen-2-yl)thiophen-2-yl)-3,5-bis(5-(5-hexylthiophen-2-yl)thiophen-2-yl)thiophene] as a HTM for an inorganic CsPbBr₃ PSC. A maximum PCE of 9.32% is obtained for the BT-BTH based device; alternatively, PEDOT, PPy and PANi are also synthesized as HTMs, yielding PCEs of 8.36%, 8.32% and 7.69% in the corresponding solar cells, respectively. Besides, the long-term stability of the BT-BTH based device free of any encapsulation maintains 94% of the initial efficiency in 70% RH humidity over 80 days. The advantages of high efficiency, good stability and low cost of CsPbBr₃ PSCs demonstrate that small molecule BT-BTH and large conjugated polymers are promising HTMs for inorganic PSCs.

Experimental

Synthesis of PEDOT, PPy and PANi

PEDOT, PPy and PANi were synthesized by an in situ chemical oxidative polymerization method. The synthesis of PEDOT was carried out in a three-necked flask filled with high-purity nitrogen. Anhydrous FeCl₃ (8.94 g, 0.055 mol) was dissolved in 180 mL of chloroform, thiophene monomer was added slowly to the flask and the mixture was agitated for 24 h at room temperature. The molar ratio of FeCl₃ to thiophene monomer was controlled at 2.3. After polymerization, the PEDOT was rinsed with methanol, 1 M hydrochloric acid and deionized water three times. Finally, the solid powders were filtrated and dried at 60 °C for 24 h.^23,24

7.788 g of FeCl₃·6H₂O was dissolved in 58 mL of deionized water, then 1 mL of pyrrole monomer was added slowly to the above solution, and the mixture was stirred for 24 h at 5 °C. After polymerization, the PPy powders were collected, rinsed by deionized water and vacuum dried at 60 °C for 24 h.²⁵ 10 mg of PPy powders was dispersed in 1 mL of chloroform to make the precursor for depositing the HTL.

(NH₄)₂S₂O₈ aqueous solution was obtained by dissolving 2.912 g of (NH₄)₂SO₄ in 100 mL of 1 M hydrochloric acid solution. PANi was prepared by dropwise adding 20 mL of (NH₄)₂S₂O₈ solution into 20 mL of a hydrochloric acid solution of aniline at 0 °C for 3 h.²⁶ Then the PANi powders were rinsed with deionized water, ethanol, and 2 M HCl solution, and dried at 60 °C for 24 h. The FTIR spectra of BT-BTH, PEDOT, PPy and PANi are shown in Fig. S3 (ESI†).

Synthesis of BT-BTH

The synthesis routes of BT-BTH and the analysis of ¹H NMR and ¹³C NMR are described in Fig. S1 (ESI†). In detail, intermediate compound 2 was obtained according to the following method. 16.6 g of 2,2′-bithiophene was dissolved in 150 mL of anhydrous tetrahydrofuran (THF), and 44 mL of n-butyl lithium (n-BuLi) and 18.15 g of bromohexane were slowly added into the mixture under an argon atmosphere. After agitating for 10 h, the reactant was cooled down and then extracted by ether. Finally, the mixture was dried and distilled under reduced pressure to obtain a colorless liquid after rinsing three times with saturated NaHCO₃ and NaCl solution. Under the same conditions, compound 2 was treated with n-BuLi, followed by the addition of tri-n-butyltin chloride, which afforded the compound 3 in 80% yield. Formation of the compound 5 was achieved by the bromination reaction with 2,2′-bithiophene and chloroform solution of HOAC. The coupling reaction of compound 3 and compound 5 yielded the target molecule BT-BTH in 70% yield.^27,28

Preparation of a TiO₂ anode

Fluorine-doped tin oxide (FTO) glass was etched by Zn powders and HCl for a desirable strip pattern and further rinsed with acetone, ethanol, and deionized water. The compact TiO₂ (c-TiO₂) layer was deposited on FTO glass by spin-coating an ethanol solution of titanium isopropoxide (0.5 M) and diethanol amine (0.5 M) at 7000 rpm for 30 s and annealing in air at 500 °C for 2 h. The mesoscopic TiO₂ (m-TiO₂) layer was then deposited by spin-coating a TiO₂ paste at 2000 rpm for 30 s and annealed in air at 450 °C for 30 min. TiO₂ paste was prepared according to our previous work.²⁹ Then the m-TiO₂ film was immersed in an aqueous solution of 0.04 M TiCl₄ at 70 °C for 30 min, cleaned with deionized water and ethanol, and finally annealed at 450 °C for another 30 min.

Assembly of solar cells

An N,N-dimethylformamide (DMF) solution of 1.0 M PbBr₂ was spin-coated onto the m-TiO₂ layer at 2000 rmp for 30 s, followed by heating at 80 °C for 30 min. Then 90 μL of 0.07 M CsBr methanol solution was coated onto FTO/c-TiO₂/m-TiO₂/PbBr₂ at 2000 rmp for 30 s and heated at 250 °C for 5 min. The CsBr was repeatedly spin-coated four times to form a CsPbBr₃ layer.^30,31 The HTM solution in chlorobenzene was spin-coated at 2000 rmp for 30 s and then heated at 80 °C for 10 min. Finally, the carbon back electrode was deposited on the CsPbBr₃ layer by doctor-blade coating carbon ink and heating at 90 °C for 10 min.

Photovoltaic measurements and characterization

The current density–voltage (J–V) curves were measured using a solar simulator (Newport, Oriel Class A, 91195A) under AM 1.5G simulated solar illumination (100 mW cm⁻², calibrated by a standard silicon solar cell). The photoelectric performances of each device were measured at least ten times to control the experimental error within ±5%. The crystal structure of a perovskite material was tested by X-ray diffraction (PHILIPSPW1800 diffractometer with Cu-Kα radiation), and the morphologies of the perovskite film and solar cell were characterized by field-emission scanning electron microscopy (FESEM, Japan Hitachi field emission S4800). The steady-state photoluminescence was recorded by an FLS920 all functional fluorescence spectrometer. The incident-photo-to-current conversion efficiency (IPCE) was characterized by a power source (Newport 300 W xenon lamp, 66920) with a monochromator (Newport Cornerstone 260) in the wavelength range of 300–1000 nm. The absorption spectra were recorded on an ultraviolet-visible spectrophotometer (UV-3600, Shimadzu) at room temperature. The time-resolved PL (TRPL) spectra were obtained on a time-resolved fluorescence spectrometer (Horiba Jobin Yvon, FL).

Results and discussion

Fig. 1a represents the preparation routes for an all-inorganic PSC device. The CsPbBr₃ perovskite layer is prepared by using a multi-step solution-processable method developed by our group: one layer of PbBr₂ and sequentially four successive layers of CsBr are spin-coated. Due to the large concentration difference between saturated PbBr₂ solution in DMF and CsBr methanol solution, the state-of-the-art one-step or two-step methods are inapplicable to make a high-purity CsPbBr₃ layer. The excessive PbBr₂ can be balanced by multiple CsBr₂ layers, resulting in a complete transformation of PbBr₂ and CsBr into perovskite-structured CsPbBr₃ phase (Fig. S2, ESI†). The stability of the perovskite is obviously promoted thanks to removal of all organic components. Fig. 1b shows a scanning electron microscopy (SEM) image of an inorganic CsPbBr₃ layer, which depicts a dense and uniform surface with a grain size of around 700 nm. HTMs are spin-coated onto the CsPbBr₃ perovskite layer to modify the CsPbBr₃/carbon interface and the corresponding SEM images of CsPbBr₃ film covered with BT-BTH, PEDOT, PPy, and PANi HTMs are shown in Fig. 1c–f. BT-BTH, PEDOT, and PPy materials have a larger grain size and higher surface coverage compared with PANi, contributing to facile charge transport and reduced energy loss at the interface.

Fig. 1 (a) The diagrammatic sketch for making TiO₂, PbBr₂, CsPbBr₃ and HTM layers. (b) Top-view SEM image of the CsPbBr₃ layer. Top-view SEM pictures of CsPbBr₃ film covered with (c) BT-BTH, (d) PEDOT, (e) PPy and (f) PANi. Scale bar: 1 μm.

The inorganic PSC has a device configuration of FTO/c-TiO₂/m-TiO₂/CsPbBr₃/HTL/carbon, as shown in Fig. 2a. The cross-sectional SEM images in Fig. 2b and Fig. S4 (ESI†) indicate a typical multi-layered structure with average thicknesses of 200 nm, 400 nm and 15 μm for c-TiO₂/m-TiO₂, CsPbBr₃ and the carbon electrode (Fig. S5, ESI†), respectively. Furthermore, the rough surface for either FTO, c-TiO₂/m-TiO₂ or CsPbBr₃ is expected to increase the adhesion of the next layer and to reduce the charge loss at cell interfaces. The fluctuant CsPbBr₃ grains lead to a rough HTM surface, but the thickness of the HTL in the valley or peak is almost the same. The homogeneous distribution of HTMs may maximize the hole extraction from CsPbBr₃ film. The J–V characteristics of the PSCs with different HTMs are measured under simulated AM 1.5G solar illumination. The optimal concentration is 3 mg mL⁻¹ by optimization of the solution concentration of HTMs (Fig. S6 and S7, ESI†). Fig. 2c displays J–V curves of all-inorganic CsPbBr₃ PSCs under a light intensity of 100 mW cm⁻² and the photovoltaic data including short-circuit density (J_sc), open-circuit voltage (V_oc), fill factor (FF), and PCE are summarized in Table 1. The HTM-free device achieves a PCE of 6.10% (J_sc = 5.93 mA cm⁻², V_oc = 1.33 V, FF = 77.5%), which is comparable to that reported by Jin's group.¹² After adding a HTM between the CsPbBr₃ layer and carbon electrode, the photovoltaic data are significantly improved. A maximum PCE of 9.32% is obtained for the BT-BTH based CsPbBr₃ PSC and the related parameters of J_sc, V_oc, and FF are 7.56 mA cm⁻², 1.50 V, and 81.9%, respectively. Moreover, the CsPbBr₃ PSCs tailored with PEDOT, PPy and PANi HTMs present PCEs of 8.36%, 8.33% and 7.69%, respectively. The improved PCE by setting the HTL at the perovskite/carbon interface may be attributed to the reduced electron–hole recombination and therefore improved charge extraction, which will be discussed in the next section.

Fig. 2 (a) Illustration of an inorganic PSC device. (b) Cross-sectional SEM image of a multilayer structure with FTO/TiO₂/CsPbBr₃/HTM configuration. (c) J–V plots and (d) IPCE spectra of these inorganic PSCs.

Table 1 Photovoltaic parameters of inorganic PSCs with different HTMs. PCE: power conversion efficiency; J_sc: short-circuit current density; V_oc: open-voltage; FF: fill factor. All the values in brackets represent their average values

Devices with HTMs	V oc (V)	J sc (mA cm^−2)	FF (%)	PCE (%)
BT-BTH	1.50 (1.51)	7.56 (7.60)	81.9 (80.0)	9.32 (9.25)
PEDOT	1.44 (1.45)	7.56 (7.55)	76.7 (76.1)	8.36 (8.48)
PPy	1.44 (1.42)	7.37 (7.30)	78.6 (77.5)	8.33 (8.30)
PANi	1.42 (1.40)	7.33 (7.30)	73.8 (73.2)	7.69 (7.89)
HTM-free	1.33 (1.35)	5.93 (5.98)	77.5 (75.4)	6.10 (6.37)

---

IPCE spectra in Fig. 2d are performed to cross-check the photoelectric behaviours of these inorganic CsPbBr₃ PSCs. It is noticeable that these inorganic CsPbBr₃ PSCs exhibit a narrower light response in a wavelength range of 325–550 nm compared with the state-of-the-art organic–inorganic PSCs.^15,32–35 In the current work, the lower PCE of CsPbBr₃ solar cells is due to a large bandgap (2.3 eV) of CsPbBr₃. The maximal IPCE value of a pristine device is around 78.3%, and it upgrades to 88.3%, 86.6%, 85.6% and 84.2% for BT-BTH, PEDOT, PPy and PANi based PSCs, respectively. The integrated J_sc values are 7.48, 7.38, 7.25, 7.21 and 5.96 mA cm⁻² for BT-BTH, PEDOT, PPy, and PANi based PSCs and the pristine device, which is consistent with the result of the J–V characteristics tests (Fig. 2c).

Fig. S8a (ESI†) shows the PCE distribution of 10 random devices tailored with different HTMs. It can be concluded that all these inorganic PSCs have high reproducibility within ±5% and the BT-BTH based PSC exhibits reasonable performance distribution and high reproducibility. Steady-state photocurrent outputs are monitored at the point of maximum power for an illumination period of 100 s, as shown in Fig. S8b (ESI†). The BT-BTH, PEDOT and pristine PSCs achieve stabilized PCEs of 9.1%, 8.2% and 6.0%, respectively. The differences between PCEs obtained from steady-state and dynamic-state characterization are mainly attributed to the hysteresis effect in the devices.

In order to investigate the recombination reaction of photo-generated charges at the CsPbBr₃/HTL interface, we study the variation of J_sc and V_oc for inorganic PSCs with and without HTMs as a function of the graded-regulated incident light intensity, as shown in Fig. 3a and b. The interior charge recombination mechanism can be understood according to the following equation: J_sc ∝ I^α (α ≤ 1),³⁶ where I represents light intensity and α is an exponential factor to evaluate charge recombination. An α value closer to 1 means lower levels of charge recombination or limited space charge effects. According to the fitting result (Fig. 3a), the α value of the pristine PSC is 0.902, while it increases to 0.994, 0.948, 0.927, and 0.920 for the BT-BTH, PEDOT, PPy, and PANi based PSCs, respectively. This suggests a reduced radiative recombination owing to a higher hole extraction ability of small molecular BT-BTH. Additionally, Fig. 3b displays the dependence of V_oc upon light intensity of the PSC devices. The relationship between V_oc and light intensity follows V_oc = εkT ln(I)/q + constant,³⁷ where ε is the ideality factor, k is the Boltzmann constant, T is the absolute temperature, and q is elementary charge. According to the Shockley–Read–Hall recombination mechanism, the slopes of the V_oc–I curves represent the degree of monomolecular recombination. The slope value is 1.848 kT/e for the pristine PSC, and it reduced to 1.479 kT/e, 1.576 kT/e, 1.658 kT/e, and 1.678 kT/e for the BT-BTH, PEDOT, PPy, and PANi based PSCs, respectively. Both investigations on J_sc and V_oc demonstrate a suppressed charge recombination by employing conductive polymers as HTMs for inorganic CsPbBr₃ PSC applications.

Fig. 3 The plots of (a) J_sc and (b) V_ocversus light intensity. (c) Diagram of energy level alignments for inorganic CsPbBr₃ PSCs. (d) TRPL decay curves for HTM-free and HTM tailored PSC devices.

The diagram of energy level alignments for the inorganic CsPbBr₃ PSC is shown in Fig. 3c. The corresponding HOMO energy level values are calculated according to cyclic voltammetry curves (Fig. S9, ESI†). And specific values are determined to be −5.44, −5.27, −5.23 and −5.20 eV for BT-BTH, PEDOT, PPy and PANi, respectively. Under sunlight illumination, the perovskite layer absorbs photons to produce electrons and holes. Photo-induced electrons transfer from their conduction band (CB) to the CB of TiO₂, while holes are collected by the carbon electrode. However, the high energy level difference (ΔE = 0.6 eV) at the CsPbBr₃/carbon interface results in serious electron–hole recombination.³⁸ HTMs can effectively reduce the energy level difference by setting an intermediate energy band between CsPbBr₃ and carbon. BT-BTH has the lowest energy level difference from CsPbBr₃, therefore it can best inhibit charge recombination and accelerate holes extraction.

Steady-state photoluminescence (PL) and the time-resolved PL (TRPL) spectra are characterized to investigate the charge-transporting performance. PL spectra of the inorganic PSC with and without polymeric HTMs are shown in Fig. S10 (ESI†). All films exhibit similar characteristic peaks at 535 nm, which is consistent with previous reports.^39–41 It should be noted that polymeric HTMs show more efficient PL quenching than the pristine PSC, especially in the case of the BT-BTH HTM tailored solar cell. The intensity of PL decreases in the following order: pristine > PANi > PPy > PEDOT > BT-BTH. The smallest PL intensity for BT-BTH means an optimal hole extraction ability. Besides, the values of electron lifetime are extracted from TRPL, as shown in Fig. 3d, and the FTO/TiO₂/CsPbBr₃/BT-BTH film presents the longest electron lifetime of 2.22 ns, which indicates the most effective hole-injection rate. Furthermore, the space-charge limited current (SCLC) tests (Fig. S11, ESI†) also indicate an increased hole mobility. All of these results manifest the promoting effect of introducing BT-BTH, PEDOT, PPy, and PANi in all-inorganic CsPbBr₃ PSCs in accelerating charge extraction and therefore enhancing photovoltaic performances.

High efficiency,^42,43 improved stability^44,45 and low cost are three criteria to evaluate the practical applications of a new-type solar cell. The cell efficiency of 9.32% is in a very high level in comparison with the highest record for a CsPbBr₃ PSC reported in our previous literature.⁴⁶ Using the solution-processable method, the cost of PSCs is regarded as half that of commercial silicon solar cells, therefore the stability of these CsPbBr₃ PSCs seems to be crucial to determine their potential commercialization. Therefore, the humidity tolerances of the BT-BTH and PEDOT based inorganic PSCs were investigated to assess their applicability. The devices without any encapsulation are measured by storing in an atmosphere of 70% RH in air at 20 °C. Fig. 4 presents the normalized PCE, J_sc, V_oc, and FF as a function of exposure time. Both devices are relatively stable and the PCE of the BT-BTH based device still maintains 94% of the initial efficiency over 80 days compared with 92% for the pristine device. The improved stability along with high solar cell efficiency demonstrates polymeric HTMs to be promising for high-performance CsPbBr₃ PSC platforms.

Fig. 4 The normalized stability for (a) PCE, (b) J_sc, (c) V_oc and (d) FF for the HTM-free solar cell and the PSCs with BT-BTH, PEDOT, PPy and PANi as HTMs. The stability of these CsPbBr₃ PSCs free of encapsulation is measured at a temperature of 20 °C and 70% relative humidity.

Conclusions

In summary, we present the feasibility of using either small molecular BT-BTH or conjugated PEDOT, PPy, and PANi as an HTM material for inorganic CsPbBr₃ PSCs incorporating HTLs to reduce charge-hole recombination at the CsPbBr₃/carbon interface for high-performance. The BT-BTH based device achieves a maximum PCE as high as 9.32% under one sun illumination, which is higher than 8.36%, 8.32%, 7.69%, and 6.10% for PEDOT, PPy, PANi and HTM-free inorganic PSCs, respectively. Detailed study demonstrates a reduced electron–hole recombination and improved charge extraction at cell interfaces. Moreover, the unencapsulated PSCs tailored with polymeric HTMs present improved humidity tolerance in high humidity over 80 days. These cost-effective and facile HTMs provide new opportunities of increasing the PCE outputs of inorganic CsPbBr₃ PSCs with sacrificing long-term stability.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge financial support from the National Natural Science Foundation of China (21503202 and 61774139) and the Fundamental Research Funds for the Central Universities (11618409).

Notes and references

1. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
2. H. Zhang, J. Mao, H. X. He, D. Zhang, H. L. Zhu, F. X. Xie, K. S. Wong, M. Grätzel and W. C. H. Choy, Adv. Energy Mater., 2015, 5, 1501354.
3. H. S. Li, Y. S. Li, Y. M. Li, J. J. Shi, H. Y. Zhang, X. Xu, J. H. Wu, H. J. Wu, Y. H. Luo, D. M. Li and Q. B. Meng, Nano Energy, 2017, 42, 222–231.
4. H. S. Kim, C. R. Lee, J. H. Im, K. B. Lee, T. Moehl, A. Marchioro, S. J. Moon, R. Humphry-Baker, J. H. Yum, J. E. Moser, M. Grätzel and N. G. Park, Sci. Rep., 2012, 2, 591.
5. NREL Chart http://www.nrel.gov/pv/assets/images/efficiency-chart.png.
6. Q. W. Han, Y. S. Bai, J. Liu, K. Z. Du, T. Y. Li, D. Ji, Y. H. Zhou, C. Y. Cao, D. Shin, J. Ding, A. D. Franklin, J. T. Glass, J. S. Hu, M. J. Therien, J. Liu and D. Mitzi, Energy Environ. Sci., 2017, 10, 2365–2371.
7. E. J. Juarezperez, Z. Hawash, S. R. Raga, L. K. Ono and Y. B. Qi, Energy Environ. Sci., 2016, 9, 3406–3410.
8. Z. Wang, Z. J. Shi, T. T. Li, Y. H. Chen and W. Huang, Angew. Chem., Int. Ed., 2017, 56, 1190–1212.
9. Y. K. Ren, X. H. Ding, Y. H. Wu, J. Zhu, T. Hayat, A. Alsaedi, Y. F. Xu, Z. Q. Li, S. F. Yang and S. Y. Dai, J. Mater. Chem. A, 2017, 5, 20327–20333.
10. Y. K. Ren, X. Q. Shi, X. H. Ding, J. Zhu, T. Hayat, A. Alsaedi, Z. Q. Li, X. X. Xu, S. F. Yang and S. Y. Dai, Inorg. Chem. Front., 2018, 5, 348–353.
11. J. L. Duan, D. W. Dou, Y. Y. Zhao, Y. D. Wang, X. Y. Yang, H. W. Yang, B. L. He and Q. W. Tang, Mater. Today Energy, 2018, 10, 146–152.
12. J. Liang, C. X. Wang, Y. R. Wang, Z. R. Xu, Z. P. Lu, Y. Ma, H. F. Zhu, Y. Hu, C. C. Xiao, X. Yi, G. Y. Zhu, H. L. Lv, L. B. Ma, T. Chen, Z. X. Tie, Z. Jin and J. Liu, J. Am. Chem. Soc., 2016, 138, 15829–15832.
13. J. Ding, Y. Y. Zhao, J. L. Duan, B. L. He and Q. W. Tang, ChemSusChem, 2018, 11, 1–7.
14. H. Z. Xu, J. L. Duan, Y. Y. Zhao, Z. B. Jiao, B. L. He and Q. W. Tang, J. Power Sources, 2018, 399, 76–82.
15. S. Mabrouk, M. M. Zhang, Z. H. Wang, M. Liang, B. Bahrami, Y. G. Wu, J. H. Wu, Q. Q. Qiao and S. F. Yang, J. Mater. Chem. A, 2018, 6, 7950–7958.
16. S. H. Peng, T. W. Huang, G. Gollavelli and C. S. Hsu, J. Mater. Chem. C, 2017, 5, 5193–5198.
17. J. Kim, N. Park, J. S. Yun, S. Huang, M. A. Green and A. W. Y. HoBaillie, Sol. Energy Mater. Sol. Cells, 2017, 162, 41–46.
18. M. C. Jung and Y. B. Qi, Org. Electron., 2016, 31, 71–76.
19. Z. Hawash, L. K. Ono and Y. B. Qi, Adv. Mater. Interfaces, 2016, 3, 1600117.
20. S. Carli, J. P. C. Baena, G. Marianetti, N. Marchetti, M. Lessi, A. Abate, S. Caramori, M. Gratzel, F. Bellina, C. A. Bignozzi and A. Hagfeldt, ChemSusChem, 2016, 9, 657–661.
21. D. Q. Bi, B. Xu, P. Gao, L. C. Sun, M. Graetzel and A. Hagfeldt, Nano Energy, 2016, 23, 138–144.
22. P. Ananthajothi and P. Venkatachalam, J. Solid State Electrochem., 2016, 20, 2633–2642.
23. F. Wu, J. Z. Cheng, R. J. Chen, S. X. Wu, L. Li, S. Chen and T. Zhao, J. Phys. Chem. C, 2011, 115, 6057–6063.
24. E. Eren, E. Aslan and A. U. Oksuz, Polym. Eng. Sci., 2015, 54, 2632–2640.
25. M. A. Garakani, S. Abouali, J. Cui and J. K. Kim, Mater. Chem. Front., 2018, 2, 1481–1488.
26. J. Yang, Y. Liu, S. Liu, L. Li, C. Zhang and T. X. Liu, Mater. Chem. Front., 2017, 1, 251–268.
27. S. X. Sun, Y. Huo, M. M. Li, X. W. Hu, H. J. Zhang, Y. W. Zhang, Y. D. Zhang, X. L. Cheng, Z. F. Shi, X. Gong, Y. S. Chen and H. L. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 19914–19922.
28. G. Gong, N. Zhao, D. B. Ni, J. Y. Chen, Y. Shen, M. K. Wang and G. L. Tu, J. Mater. Chem. A, 2016, 4, 3661–3666.
29. Q. W. Tang, J. Wang, B. L. He and P. Z. Yang, Nano Energy, 2017, 33, 266–271.
30. J. L. Duan, T. Y. Hu, Y. Y. Zhao, B. L. He and Q. W. Tang, Angew. Chem., Int. Ed., 2018, 57, 5746–5749.
31. J. L. Duan, Y. Y. Zhao, B. L. He and Q. W. Tang, Angew. Chem., Int. Ed., 2018, 57, 3787–3791.
32. B. B. Zong, W. Y. Fu, H. J. Liu, L. W. Huang, H. Bala, X. D. Wang, G. Sun, J. L. Cao and Z. Y. Zhang, J. Alloys Compd., 2018, 748, 1006–1012.
33. Z. Y. Liu, B. Sun, X. Y. Liu, J. H. Han, H. B. Ye, Y. X. Tu, C. Chen, T. L. Shi, Z. R. Tang and G. L. Liao, J. Mater. Chem. A, 2018, 6, 7409–7419.
34. Y. Liu, H. Zhang, Y. P. Zhang, B. Xu, L. J. Liu, G. Chen, C. Im and W. J. Tian, J. Mater. Chem. A, 2018, 6, 7922–7932.
35. L. Li, Y. Q. Wang, X. L. Deng, L. H. Nie, Z. D. Cui and C. W. Shi, J. Nanosci. Nanotechnol., 2018, 18, 5095–5100.
36. K. Chen, Q. Hu, T. H. Liu, L. C. Zhao, D. Y. Luo, J. Wu, Y. F. Zhang, W. Zhang, F. Liu, T. P. Russell, R. Zhu and Q. H. Gong, Adv. Mater., 2016, 28, 10718.
37. A. Pockett, G. E. Eperon, T. Peltola, H. J. Snaith, A. Walker, L. M. Peter and P. J. Cameron, J. Phys. Chem. C, 2015, 119, 3456–3465.
38. A. Mei, X. Li, L. Liu, Y. Rong, M. Xu, M. Hu, J. Chen, Y. Yang, M. Grätzel and H. Han, Science, 2014, 345, 295–298.
39. X. S. Zhan, Z. W. Jin, J. R. Zhan, D. L. Bai, H. Bian, K. Wang, J. Sun, Q. Wang and S. F. Liu, ACS Appl. Mater. Interfaces, 2018, 10, 7145–7154.
40. Z. Y. Liu, B. Sun, X. Y. Liu, J. H. Han, H. B. Ye, T. L. Shi, Z. R. Tang and G. L. Liao, Nano-Micro Lett., 2018, 10, 34–46.
41. X. W. Chang, W. P. Li, L. Q. Zhu, H. C. Liu, H. F. Geng, S. S. Xiang, J. M. Liu and H. N. Chen, ACS Appl. Mater. Interfaces, 2016, 8, 33649–33655.
42. M. L. Cai, N. Ishida, X. Li, X. D. Yang, T. Noda, Y. Z. Wu, F. X. Xie, H. Naito, D. Fujita and L. Y. Han, Joule, 2018, 2, 296–306.
43. J. Zhang, D. L. Bai, Z. W. Jin, H. Bian, K. Wang, J. Sun, Q. Wang and S. Z. Liu, Adv. Energy Mater., 2018, 8, 1703246.
44. T. Niu, R. Munir, J. B. Li, D. Barrit, X. Zhang, H. L. Hu, Z. Yang, A. Amassian, K. Zhao and S. Z. Liu, Adv. Mater., 2018, 30, 1706576.
45. Y. Wang, T. Y. Zhang, F. Xu, Y. H. Li and Y. X. Zhao, Sol. RRL, 2018, 2, 1700180.
46. J. L. Duan, Y. Y. Zhao, X. Y. Yang, Y. D. Wang, B. L. He and Q. W. Tang, Adv. Energy Mater., 2018 DOI:10.1002/aenm.201802346.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c8qm00337h

‡ These authors made equal contributions to this work.

---