Improvement of stability of ZnO/CH₃NH₃PbI₃ bilayer by aging step for preparing high-performance perovskite solar cells under ambient conditions

Abstract

To date, most of fabricating processes of efficient perovskite solar cells (PSCs) have been carried out in high vacuum conditions or nitrogen protection atmosphere. Here we report a simple and novel strategy to fabricate efficient and stable ZnO-based PSCs. Firstly, ZnO nanoparticles were prepared by zinc acetate hydrolysis in methanol, and then the growth of methylamine lead iodide (CH₃NH₃PbI₃) crystal layer was realized by a two-step method. With aging treatment of ZnO film, high-quality ZnO/CH₃NH₃PbI₃ bilayer was obtained. Most importantly, the fabrication, measurement and store of solar cells in our processes were totally carried out under ambient conditions without a high vacuum or oxygen-free and water-free environment. A maximum efficiency of 9.4% was achieved. Moreover, the prepared devices exhibited promising stability at 30% humidity and 25 °C without any protection. We believe that the proposed strategy would push the applications of PSCs.

---

Introduction

Perovskite solar cells (PSCs) have attracted extensive international attention since first being introduced in 2009.¹ The solid development of PSCs began with the pioneering work performed by Snaith² and Park³ in 2012. From then on, so-called all-solid state PSCs have been considered as next generation photovoltaic devices due to their high power conversion efficiency (PCE) and low-cost production.^4–7 To date, a certified PCE of PSCs as high as 20.1% was demonstrated.⁸ Most recently, Han and co-workers reported high efficiency PSCs (PCE > 15%) with an aperture area > 1 cm².⁹ The PSCs with gradually improved PCE and significant breakthroughs hold great promise to solve the global energy problem.

Currently, most efficient PSCs are based on nanostructured TiO₂ photoanodes. In this typical structure, high-temperature thermal treatment process is often required to gain high-quality TiO₂ nanostructure, which could induce high manufacturing cost and damage the plastic substrate.^10,11 Comparing with TiO₂, use of ZnO as electron transport material has at least two obvious advantages: higher electron mobility and free from higher temperature sintering.¹² Exploiting ZnO as an ideal alternative to the conventional TiO₂ ought to be an effective way to avoid high-temperature process. It has been verified that ZnO with a decent crystallinity can be synthesized by using various techniques within low temperature ranges,¹³ which is well suitable for flexible solar cells.¹⁴ In addition, ZnO is a wide band gap inorganic semiconductor oxide, which has higher electronic mobility than that of TiO₂. Recently, ZnO has been widely employed as the electron transport layer (ETL) of PSCs. For example, Hagfeldt and co-workers first reported an efficient and stable spiro-MeOTAD/CH₃NH₃PbI₃/ZnO solid-state solar cell.¹⁵ The group of Park increased the PCE from 11.13% to 14.35% through ZnO nanorod as an effective charge collection system.^16,17 The group of Pauporté also reported low-temperature manufacturing high performance ZnO-based PSCs.^10,18,19 Liu and Kelly demonstrated remarkably high efficiency PSCs with a planar heterojunction structure based on ZnO electron-transport layers.²⁰ Amassian and co-workers have obtained 16.1% efficient hysteresis-free ZnO-based PSCs.²¹

Despite the significant improvement in the PCE for ZnO-based PSCs, there are still numerous challenges for commercial applications. A key challenge is to reduce the manufacturing cost because most of fabricating processes reported were carried out in a high vacuum or oxygen-free and water-free environment.²² That another important issue needs to be addressed is to improve the long-term stability of the devices under different circumstances.^23–25 Especially, in ZnO-based PSCs, it is very difficult to maintain high-quality ZnO/CH₃NH₃PbI₃ bilayer.^26,27

In this present study, we report a simple and novel strategy to fabricate efficient and stable ZnO-based PSCs. For the first time, we employed aging step of ZnO films to enhance thermal stability of perovskite films and devices. Notably, the fabrication and measurement processes of solar cells were totally carried out under ambient conditions at room temperature. We obtained PCE of up to 9.4% albeit fluorine doped tin oxide (FTO) glass was used. The prepared devices exhibited promising stability even with simple fabricating procedures under ambient conditions.

Experimental section

Preparation of ZnO nanoparticles

ZnO nanoparticles were synthesized by combining the previously reported work with some modifications.^20,28 In a typical procedure, 52 mL of KOH methanol solution (354 mM) was first added dropwise into a 100 mL of Zn(CH₃COO)₂·2H₂O methanol solution (107 mM) in a 65 °C water bath and continuously stirred for 2.5 h. Then, the products were washed with anhydrous methanol solution for several times to remove residual precursors. Finally, the prepared ZnO nanoparticles were dispersed into a 20 mL mixture solution of n-butanol–methanol–chloroform (14 : 1 : 1 volume ratio), where the content of ZnO nanoparticles is about 20 mg mL⁻¹. The obtained ZnO solution was filtered by a 0.45 μm PTFE syringe filter before spin-coating.

Fabrication of ZnO-based PSCs

FTO rigid glass substrates (1.5 cm × 2.0 cm) were etched by zinc power and 2 M HCl, and treated successively with abluent, deionized water, acetone and UV-ozone to achieve a clean surface. After that, the ZnO nanoparticles solution was spin-coated over the FTO substrate at 3000 r.p.m. for 30 s and then dried for 10 min at room temperature. This process was repeated several times to obtain an optimal thickness of ZnO film. Consecutively, the ZnO film was aged 24 h in air at room temperature. Then, solution of PbI₂ in N,N-dimethylformamide (DMF) (460 mg mL⁻¹) was spin-coated onto the surface of ZnO films at 3000 r.p.m. for 20 s. After drying the PbI₂ layer, the substrate with both ZnO and PbI₂ layers was immersed into a solution of CH₃NH₃I in 2-propanol (10 mg mL⁻¹) for 1 min, followed by thermal annealing at 80 °C for 20 min (FTO/ZnO/CH₃NH₃PbI₃). Subsequently, the hole transfer material 2,2′,7,7′-tetrakis-(N,N-di-4-methoxyphenylamino)-9,9′-spirobifluorene (Spiro-OMeTAD) was deposited on the top of FTO/ZnO/CH₃NH₃PbI₃ at 4000 r.p.m. for 30 s, where 80 mg of spiro-OMeTAD was dissolved in the mixture solution of 28.5 μL 4-tert-butypyridine, 17.5 μL lithium bis(trifluoromethane-sulfonyl)imide (Li-TFSI) (520 mg of Li-TFSI in 1 mL of acetonitrile) and 1 mL chlorobenzene. Finally, Ag counter electrode with 100 nm in thickness was deposited via vacuum thermal evaporation. In this work, the solar cells involving the structure of FTO/ZnO/CH₃NH₃PbI₃/Spiro-OMeTAD/Ag were fabricated, measured and stored under ambient conditions without any specific protection, where the humidity and temperature are about 30% and 25 °C, respectively.

Characterization

XRD patterns were collected on a Shimadzu X-ray diffractometer 6000 with Cu Kα radiation (λ = 1.5418 Å), which was used to characterize the crystal structure and phase of material. Raman spectra were obtained on a Renishaw in via confocal micro-Raman spectroscopy system. Fourier transform infrared spectra (FT-IR) were recorded on an Avatar 360 (Nicolet) instrument. The morphology of ZnO nanoparticles and film were characterized by a transmission electron microscope (TEM, FEI, Tecnai G2 F20) and a scanning electron microscopy (SEM, Hitachi S4800 HSD).

Photocurrent density–photovoltage (J–V) performance of the devices were measured using an electrochemical workstation (VersaSTAT 3, Ametek, USA), where the cells were illuminated by a 150 W xenon lamp class ABB solar simulator (94021A, Newport, USA) under an AM1.5G radiation at a calibrated intensity of 100 mW cm⁻², as standardized by a standard Si solar cell (1218, Newport, USA). Incident photon-to-electron conversion efficiency (IPCE) was recorded by a solar cell quantum efficiency measurement system (QTest Station 500D, Crowntech, USA) equipped with a power source (300 W tungsten lamp), a monochromator and a multimeter. The active area of the cell was confirmed to be 0.12 cm² through a nonreflective metal mask.

Results and discussion

So far, most of the ZnO-based PSCs devices with ideal performance have usually been structured on the indium doped tin oxide (ITO) substrate rather than the FTO, because the former involves a better transmittance.²⁹ However, the utility of ITO glass containing an energy-intensive rare element (i.e. indium) is against the principle of environmental friendly development and also raise the cost of PSCs. We herein applied the budget FTO, replacing of the most common used ITO, as the device supporting substrate to fabricate the ZnO-based PSCs under the ambient condition without any specific protection. The structure diagram and the photoelectric conversion mechanism of ZnO-based PSCs are illustrated in Fig. 1. The component units, from the bottom up, are Ag electrode, hole transport layer (HTL) Spiro-OMeTAD, light absorption layer (LAL) CH₃NH₃PbI₃, ETL ZnO and FTO glass. When sunlight irradiating from the direction of FTO, CH₃NH₃PbI₃ as the absorber will capture the transmitted photons and then produce electron–hole pairs. These photo-induced electrons and holes will transfer from ETL and HTL to the terminals (FTO glass and Ag electrode) of the device, proceeding to generate current by connecting the external circuit.

Fig. 1 Schematic diagram of PSCs prepared under ambient conditions.

In the fabrication process of ZnO-based PSCs, ZnO nanoparticles layer was first spin-coated over the transparent FTO substrate to form a dense thin film. TEM image and the corresponding size distribution of ZnO nanoparticles showed that monodisperse ZnO nanoparticles with homogeneous size were obtained (Fig. 2a and b). The uniform ZnO nanoparticles are no doubt to be beneficial for preparing a high-quality ZnO layer. The component of ZnO film was confirmed by Raman spectroscopy (Fig. 2c). It can be seen clearly that there are two prominent Raman peaks at 565 cm⁻¹ and 1086 cm⁻¹, which can be ascribed to the asymmetric multiphonons vibration of Zn sublattice and oxygen atoms.³⁰ Furthermore, the XRD pattern (Fig. 2d) was executed to analysis the crystallinity and crystal phase of ZnO nanoparticles. The characteristic diffraction peaks at 31.61°, 34.16°, 36.10°, 47.44°, 56.61°, 62.73° and 67.83° can be well indexed to (100), (002), (101), (102), (110), (103) and (112) crystal planes of the high-purity wurtzite hexagonal phase in ZnO nanoparticles.³¹ Remarkably, some other peaks were also obtained whose intensity are even stronger than that of ZnO nanoparticles. According to the reported work and the JCDPS 41-1445, these additional peaks at 26.48°, 37.65°, 51.57°, 54.61°, 61.59° and 65.49° are attributed to the (110), (200), (211), (220), (310), (301) crystal planes of SnO₂ conductive thin film coated on the glass substrate (FTO).^32,33 The broad peak at 22.06° refers to the amorphous silica. These data, taken together, demonstrated that ZnO nanoparticles with high purity and crystallinity had been synthesized successfully.

Fig. 2 (a) TEM image of ZnO nanoparticles. (b) Size distribution of ZnO nanoparticles. (c) Raman spectrum of ZnO layers on FTO glass (FTO/ZnO film). (d) XRD pattern of FTO/ZnO film.

According to the previously reported method, the freshly prepared ZnO film should be dried for 10 min at room temperature and then the PbI₂ solution was spin-coated on it.²⁰ After drying about 5 min, the substrate was immersed into a solution of CH₃NH₃I in 2-propanol for 1 min. To obtain a high-quality CH₃NH₃PbI₃ film, the thermal annealing process is extremely crucial.³⁴ Therefore, the sample composed of FTO glass, ZnO film and initialized CH₃NH₃PbI₃ was annealed at 80 °C. However, it was very difficult to achieve thermal stability of the ZnO/CH₃NH₃PbI₃ bilayer in our case. As shown in Fig. 3a, it was found that some yellow spots usually appeared in the originally expected dark brown CH₃NH₃PbI₃ layer and spread over the entire surface within a period of 45 min. This indicates that CH₃NH₃PbI₃ phase is likely to involve an undesirable thermal decomposition in the presence of ZnO layer, resulting in low efficiency and instability in PSCs. In order to confirm the decomposition of CH₃NH₃PbI₃ film, the structure of the yellow sample was investigated by XRD. The XRD pattern (Fig. 3c) indicates the presence of PbI₂, which exhibits the typical strong diffraction peak of PbI₂ at around 12.48°.³⁵ This finding is similar to the previous investigation reported by Hu et al.,³⁶ who demonstrated that CH₃NH₃PbI₃ film underwent fast thermal decomposition on ZnO-coated ITO. The origin of the decomposition of perovskite crystal may arise from the ZnO-based electron selective layer. The instability of CH₃NH₃PbI₃ film could be written:³⁷

CH₃NH₃PbI₃ ↔ CH₃NH₃I + PbI₂ (1)

Fig. 3 FT-IR spectrum obtained from (a) fresh and (b) aged FTO/ZnO film. Inserts show the corresponding digital pictures of CH₃NH₃PbI₃ layers coated on fresh (yellow) and aged (dark brown) FTO/ZnO film after thermal annealing at 80 °C for 45 min. XRD patterns are from (c) yellow and (d) dark brown samples in the inserts of (a) and (b). Top view SEM images of the optimum (e) ZnO film and (f) CH₃NH₃PbI₃ film.

This decomposition reaction will be accelerated with temperature, residual solvent and precursors. To investigate the chemical species on ZnO surface, FT-IR was employed. As shown in Fig. 3a, there are two typical vibrational modes at 1568 cm⁻¹ and 1417 cm⁻¹, the former stronger than the latter one, which are well consistent to the coupling of the C=O stretching vibrations.³⁸ Moreover, the wide band around 3400 cm⁻¹ is responsible for the hydroxyl radical. Our evidence for the presence of hydroxyl groups and acetate ligands are consistent with previous measurements made by Yang et al.,²⁶ who points out that the decomposition process of CH₃NH₃PbI₃ was promoted by residual hydroxyl groups and acetate ligands on the ZnO surface.

To address this challenge, a few attempts have been made to improve the thermal stability of CH₃NH₃PbI₃ film on ZnO. Manspeaker and co-workers introduced a restricted volume solvent annealing process to ZnO-based PSCs to improve the stability of the devices.³⁹ Song et al. reported that ZnO–SnO₂ hybrid film can suppress the thermal decomposition of CH₃NH₃PbI₃ and show better thermal stability than that of ZnO film.⁴⁰ Tsang and co-workers added a polymeric buffer layer between ZnO and perovskite film to block the direct contacting of them. This type of PSCs have a PCE of 10.2%.⁴¹ Adopting this method can effectively get rid of the decomposition of CH₃NH₃PbI₃ and further improve the stability of ZnO-based PSCs devices. Although the quality of perovskite crystal have been improved, more complex preparation techniques are necessary. For the first time, we herein utilized a novel method to solve this problem, simply aging ZnO film in air at room temperature. Our method not only remarkably improved morphology of ZnO film, but also effectively removed hydroxyl groups and residual acetate ligands on the ZnO surface. In our case, before depositing the PbI₂, the ZnO film was aged for 24 h at room temperature under the atmosphere environment. As expected, the CH₃NH₃PbI₃ layer on aged FTO/ZnO showed an excellent thermal stability. In spite of annealing for 45 min at 80 °C, the CH₃NH₃PbI₃ film still kept dark brown. To identify the crystal phase, XRD pattern of the dark brown sample was shown in Fig. 3d. These high-intensitive diffraction peaks at 13.98°, 28.32°, 31.78° are in accordance with the (110), (220), (310) crystal planes of the tetragonal phase of CH₃NH₃PbI₃.⁴² The results suggested that most of the CH₃NH₃PbI₃ phase could be survived in the annealing process. Surprisingly, there was minor PbI₂ component in the film. The FT-IR spectrum was collected from aged ZnO film to investigate the function of the aging method. As shown in Fig. 3b, we found that no any peak could be observed, which indicated that acetate and hydroxyl had been removed thoroughly. As shown in Fig. 3e, SEM image of aged ZnO film was conducted to evaluate the surface morphology. It can be seen that the ZnO film is more dense and uniform comparing to the film annealed at 70 °C for 30 min (Fig. S1†), which is beneficial for transporting electrons and then improving the performance of the devices. Fig. 3f shows the SEM image of dark brown CH₃NH₃PbI₃ film, where the typical tetragonal structure of CH₃NH₃PbI₃ crystals can be observed clearly. Combining the XRD pattern with the SEM image of dark brown CH₃NH₃PbI₃ film, we can make a conclusion that the tetragonal CH₃NH₃PbI₃ crystal with a fraction of PbI₂ can be achieved upon the aged FTO/ZnO film. The facile preparation process and improved thermal stability of ZnO/CH₃NH₃PbI₃ bilayer make our approach a new choice for highly efficient and stable PSCs.

It is well-known that the high electron mobility is one of the most important factors to introduce ZnO film as the electron selective and hole blocking layer. Also, the electron mobility of ZnO layer is closely associated with its thickness, which will determine the lifetime of the photoinduced electrons and then influence the PCE of devices.⁴³ To obtain the optimal thickness, ZnO nanoparticles were spun different times (one, two, three, four and five) onto FTO substrates step by step. Afterwards, all of the FTO/ZnO film with different ZnO layers were aged for 24 h at room temperature. Subsequently, the CH₃NH₃PbI₃ layer, spiro-OMeTAD layer and Ag counter electrode were successively introduced to constitute a series of integrated devices. To evaluate the performance of the prepared devices, the photocurrent density–photovoltage (J–V) measurements were carried out under an one sun illuminating with a calibrated intensity of 100 mW cm⁻² at ambient conditions. Fig. 4 shows the plots of photovoltaic parameters, including short-circuit current density (J_SC), open-circuit voltage (V_OC), fill factor (FF) and efficiency (PCE), as a function of the ZnO layer numbers. It is worthy to note that these photovoltaic parameters obtained from J–V curves of ten devices. We can see that the FF of ZnO-based PSCs increases at beginning and then fluctuates around a constant level. Interestingly, the results of J_SC and PEC augment linearly with the increase of ZnO layers from one to four, and drop abruptly while becoming thicker. As for V_OC, three and more times of spin-coating of ZnO nanoparticles endow a maximum value. Taking into account all of these above-mentioned factors, we can ascertain that the ETL containing a four-layer ZnO film will result in the most effective electrons transport and the lowest interfacial recombination among them. The devices with four-layer ZnO nanoparticles show an averaged PCE of 9.2% along with high J_SC of 18.34 mA cm⁻², V_OC of 0.84 V and suitable FF of 0.59. In theory, thin ETL layer is beneficial to reducing interfacial damping and promoting the electrons propagation efficiency. However, ZnO layer could not cover all of the protuberances on the rough surface of FTO substrate when spin-coating times are limited.⁴⁴ In this case, a large amount of pinholes will present in the ZnO layer, which could become active sites of recombination of electrons and holes, and immensely impair the performance of the devices. Conversely, too thick ZnO film will raise electron transport resistance, leading to the recombination of carriers.³ Thus, adjusting and confirming the spin-coating times of ZnO nanoparticle is important to obtain ideal ZnO-based PSCs.

Fig. 4 Changing of (a) J_SC, (b) V_OC, (c) PCE, and (d) FF of ten devices with the increase of ZnO layer numbers.

Afterwards, the devices fabrication was based on the optimized structure of FTO/ZnO/CH₃NH₃PbI₃/Spiro-OMeTAD/Ag. Fig. 5a and b show the cross sectional view of FTO/ZnO/CH₃NH₃PbI₃ film at low and high magnifications, respectively. It can be seen that well-aligned boundary lines among these different kinds of layers, in which the thickness of ZnO and CH₃NH₃PbI₃ films are about 74 nm and 368 nm. More careful observation of the junction between ZnO and CH₃NH₃PbI₃ layers can verify the well contact of them. The J–V curve of the best device was extracted to show in Fig. 6a. The device displays an excellent efficiency of 9.4%, whose J_SC, V_OC and FF were determined as 18.35 mA cm⁻², −0.86 V and 0.6, respectively. As to the solar cell preparing and measuring under ambient conditions without glovebox, these data ought to represent a relatively high performance. Moreover, in order to further evaluate the photoelectricity characteristics of the solar cell under different wavelengths, we also tested the IPCE spectrum. As shown in Fig. 6b, we can clearly observe that the external quantum efficiency value of the device are higher than 60% in the wavelength range from 380 to 740 nm and the plateau of that can even reach 73%. The integrated current density is 16.23 mA cm⁻², which approaches to the measured value of 18.35 mA cm⁻² through J–V curve. This high IPCE corresponding to a decent J_SC for ZnO-based PSCs indicates that the ZnO film prepared via free heating process has a good light-harvesting and photo-induced electrons collecting capacity. Such a high electron collecting capacity should be first attributed to the excellent electron mobility of ZnO in nature. In addition to the intrinsic property of ZnO, the facile aging method without heating and the presence of PbI₂ phase in the perovskite layer would also play positive roles for the PCE of device. Comparing with the high-temperature procedure,²⁶ the aging method at room temperature can effectively avoid cracking the continuous ZnO film to ensure the ability of electrons transporting. As mentioned above, a small amount of PbI₂ phase existing in the perovskite grain boundaries will form CH₃NH₃PbI₃/PbI₂ interfaces, which could speed up the transmission and reduce the carriers recombination on the surface of absorption layer and interface between ZnO and perovskite.^45,46 On one hand, the existence of passivated CH₃NH₃PbI₃/PbI₂ interfaces would form new p–i–n structures in PSCs, reducing the carriers recombination. On the other hand, CH₃NH₃PbI₃/PbI₂ interfaces in absorption layer can also improve the compactness of perovskite film, as well as strengthen the connection between perovskite absorption layer and electrons/holes transport layer.⁴⁷ It is anticipated that the high-performance ZnO-based PSCs can be obtained by further control of the amount and morphology of PbI₂.

Fig. 5 Cross sectional view of FTO/ZnO/CH₃NH₃PbI₃ film at (a) low and (b) high magnifications.

Fig. 6 (a) J–V curve and (b) IPCE spectrum of the optimal ZnO-based PSCs.

To date, the poorly environmental stability of PSCs is still the leading obstacle in practical applications.²⁵ It was previously reported that the ZnO-based PSCs are pretty vulnerable at atmosphere conditions.¹⁹ It is worth noting that, the ZnO-based PSCs via aging treatment in this work show an excellent stability. We exposed ZnO-based PSCs, without any specific protection, to the ambient air for 35 days and measured the J–V curves every seven days (25 °C and 30% humidity). The plots of photovoltaic parameters as a function of the storage period are shown in Fig. 7. These photovoltaic parameters are also from ten devices. We can see that the J_SC slowly decrease from 18.01 mA cm⁻² to 15.21 mA cm⁻² within 35 days. The FF almost keeps a stable level while the V_OC shows a slight decrease with the extension of storage time. Thus the PCE of devices demonstrate a favorable level after 35 days, which still remain 81% of the original level. Meanwhile, we also carried out the XRD pattern of the ZnO/CH₃NH₃PbI₃ bilayer stored for 35 days under the same conditions (Fig. S2†). The result shows that the diffraction peaks can well match with those of new prepared ZnO/CH₃NH₃PbI₃ film. The peak intensity of PbI₂ in ZnO/CH₃NH₃PbI₃ bilayer stored in air conditions for 35 days still keeps a relatively low level. All of these results suggest that the crystallographic structure of CH₃NH₃PbI₃ layer deposited on aged ZnO film have a better stability. We suppose that such a high durability is mostly due to the absence of residual hydroxyl groups and acetate ligands in the aged ZnO film, which can effectively restrain the decomposition of CH₃NH₃PbI₃. Moreover, it has been reported that the passivation of PbI₂ for CH₃NH₃PbI₃ is also helpful to improve the stability of devices.^14,48,49 Combined with the high efficiency, these results suggest that our strategies provide a promising route to fabricate environmentally stable and high-performance ZnO-based PSCs under ambient conditions.

Fig. 7 Changing of (a) J_SC, (b) V_OC, (c) PCE, and (d) FF of ten devices with the extension of storage period. The devices were stored in ambient conditions with about 30% humidity and 25 °C.

Conclusion

In this work, we found that the residual hydroxyl groups and acetate ligands existing in the ZnO film is the main cause of accelerating CH₃NH₃PbI₃ decomposition during the annealing process. Here, we have successfully removed undesirable chemical species in ZnO layers by a simple aging method. Notably, the aging method was carried out at room temperature and could well keep the compactness and continuity of ZnO film. This is in favour of accelerating electrons transmission and reducing the carriers recombination. Moreover, integrating the passivation role of PbI₂, the PCE and environmental stability of ZnO-based PSCs were substantially improved. It should be noted that all of processes, such as fabrication, measurement and storage, were conducted under ambient conditions without glovebox. We believe that this work ought to take a solid step in practical applications.

Acknowledgements

This work is supported by Specialized Research Fund for the Doctoral Program of Higher Education (20122302110043), the State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (2016DS07), Key Laboratory of Functional Inorganic Material Chemistry (Heilongjiang University) and Ministry of Education.

References

1. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
2. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647.
3. 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. Graetzel and N.-G. Park, Sci. Rep., 2012, 2, 591.
4. H. S. Jung and N. G. Park, Small, 2015, 11, 10–25.
5. T. Salim, S. Sun, Y. Abe, A. Krishna, A. C. Grimsdale and Y. M. Lam, J. Mater. Chem. A, 2015, 3, 8943–8969.
6. S. Kazim, M. K. Nazeeruddin, M. Grätzel and S. Ahmad, Angew. Chem., Int. Ed., 2014, 53, 2812–2824.
7. H. J. Snaith, J. Phys. Chem. Lett., 2013, 4, 3623–3630.
8. W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo and S. I. Seok, Science, 2015, 348, 1234–1237.
9. W. Chen, Y. Wu, Y. Yue, J. Liu, W. Zhang, X. Yang, H. Chen, E. Bi, I. Ashraful and M. Grätzel, Science, 2015, 350, 944–948.
10. J. Zhang, E. J. Juárez-Pérez, I. Mora-Seró, B. Viana and T. Pauporté, J. Mater. Chem. A, 2015, 3, 4909–4915.
11. X. Bao, Y. Wang, Q. Zhu, N. Wang, D. Zhu, J. Wang, A. Yang and R. Yang, J. Power Sources, 2015, 297, 53–58.
12. Q. Zhang, C. S. Dandeneau, X. Zhou and G. Cao, Adv. Mater., 2009, 21, 4087–4108.
13. I. Gonzalez-Valls and M. Lira-Cantu, Energy Environ. Sci., 2009, 2, 19–34.
14. H. Liu, Z. Huang, S. Wei, L. Zheng, L. Xiao and Q. Gong, Nanoscale, 2016, 8, 6209–6221.
15. D. Bi, G. Boschloo, S. Schwarzmüller, L. Yang, E. M. Johansson and A. Hagfeldt, Nanoscale, 2013, 5, 11686–11691.
16. D. Y. Son, J. H. Im, H. S. Kim and N. G. Park, J. Phys. Chem. C, 2014, 118, 16567–16573.
17. D. Y. Son, K. H. Bae, H. S. Kim and N. G. Park, J. Phys. Chem. C, 2015, 119, 10321–10328.
18. J. Zhang, P. Barboux and T. Pauporté, Adv. Energy Mater., 2014, 4, 1–8.
19. J. Zhang and T. Pauporté, J. Phys. Chem. C, 2015, 119, 14919–14928.
20. D. Liu and T. L. Kelly, Nat. Photonics, 2014, 8, 133–138.
21. K. Mahmood, B. S. Swain and A. Amassian, Adv. Energy Mater., 2015, 5, 1–11.
22. Y. Jin and G. Chumanov, ACS Appl. Mater. Interfaces, 2015, 7, 12015–12021.
23. G. Niu, X. Guo and L. Wang, J. Mater. Chem. A, 2015, 3, 8970–8980.
24. T. Leijtens, G. E. Eperon, N. K. Noel, S. N. Habisreutinger, A. Petrozza and H. J. Snaith, Adv. Energy Mater., 2015, 5, 1–23.
25. T. A. Berhe, W. N. Su, C. H. Chen, C. J. Pan, J. H. Cheng, H. M. Chen, M. C. Tsai, L. Y. Chen, A. A. Dubale and B. J. Hwang, Energy Environ. Sci., 2016, 9, 323–356.
26. J. L. Yang, B. D. Siempelkamp, E. Mosconi, F. De Angelis and T. L. Kelly, Chem. Mater., 2015, 27, 4229–4236.
27. J. Liu, Y. Shirai, X. Yang, Y. Yue, W. Chen, Y. Wu, A. Islam and L. Han, Adv. Mater., 2015, 27, 4918–4923.
28. W. J. Beek, M. M. Wienk, M. Kemerink, X. Yang and R. A. Janssen, J. Phys. Chem. B, 2005, 109, 9505–9516.
29. J. Gong, S. B. Darling and F. You, Energy Environ. Sci., 2015, 8, 1953–1968.
30. R. Zhang, P. G. Yin, N. Wang and L. Guo, Solid State Sci., 2009, 11, 865–869.
31. A. K. Zak, W. A. Majid, M. E. Abrishami and R. Yousefi, Solid State Sci., 2011, 13, 251–256.
32. D. Zhao, W. Ke, C. R. Grice, A. J. Cimaroli, X. Tan, M. Yang, R. W. Collins, H. Zhang, K. Zhu and Y. Yan, Nano Energy, 2016, 19, 88–97.
33. J. Song, E. Zheng, J. Bian, X. Wang, W. Tian, Y. Sanehira and T. Miyasaka, J. Mater. Chem. A, 2015, 3, 10837–10844.
34. D. Liu, L. Wu, C. Li, S. Ren, J. Zhang, W. Li and L. Feng, ACS Appl. Mater. Interfaces, 2015, 7, 16330–16337.
35. X. Wang, Z. Li, W. Xu, S. A. Kulkarni, S. K. Batabyal, S. Zhang, A. Cao and L. H. Wong, Nano Energy, 2015, 11, 728–735.
36. Q. Hu, J. Wu, C. Jiang, T. Liu, X. Que, R. Zhu and Q. Gong, ACS Nano, 2014, 8, 10161–10167.
37. G. Niu, W. Li, F. Meng, L. Wang, H. Dong and Y. Qiu, J. Mater. Chem. A, 2014, 2, 705–710.
38. X. Cheng, H. Zhao, L. Huo, S. Gao and J. Zhao, Sens. Actuators, B, 2004, 102, 248–252.
39. C. Manspeaker, P. Scruggs, J. Preiss, D. A. Lyashenko and A. Zakhidov, J. Phys. Chem. C, 2016, 120, 6377–6382.
40. J. Song, E. Zheng, X. Wang, W. Tian and T. Miyasaka, Sol. Energy Mater. Sol. Cells, 2016, 144, 623–630.
41. Y. H. Cheng, Q. D. Yang, J. Y. Xiao, Q. F. Xue, H. W. Li, Z. Q. Guan, H. L. Yip and S. W. Tsang, ACS Appl. Mater. Interfaces, 2015, 7, 19986–19993.
42. J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316–319.
43. Y. Xiao, G. Han, Y. Chang, Y. Zhang, Y. Li and M. Li, J. Power Sources, 2015, 286, 118–123.
44. K. M. Lee, S. H. Chang, K. H. Wang, C. M. Chang, H. M. Cheng, C. C. Kei, Z. L. Tseng and C. G. Wu, Sol. Energy, 2015, 120, 117–122.
45. T. Supasai, N. Rujisamphan, K. Ullrich, A. Chemseddine and T. Dittrich, Appl. Phys. Lett., 2013, 103, 1–3.
46. Y. H. Lee, J. Luo, R. Humphry-Baker, P. Gao, M. Grätzel and M. K. Nazeeruddin, Adv. Funct. Mater., 2015, 25, 3925–3933.
47. Q. Chen, H. Zhou, T. B. Song, S. Luo, Z. Hong, H. S. Duan, L. Dou, Y. Liu and Y. Yang, Nano Lett., 2014, 14, 4158–4163.
48. J. Song, J. Bian, E. Zheng, X. Wang, W. Tian and T. Miyasaka, Chem. Lett., 2015, 44, 610–612.
49. E. Zheng, X. F. Wang, J. Song, L. Yan, W. Tian and T. Miyasaka, ACS Appl. Mater. Interfaces, 2015, 7, 18156–18162.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1. SEM image of ZnO film annealed at 70 °C for 30 min; Fig. S2. XRD patterns of aged ZnO/CH₃NH₃PbI₃ bilayer. The red line is collected from fresh sample and the black line refers to the one stored for 35 days under ambient conditions (25 °C and 30% humidity). See DOI: 10.1039/c6ra10072d

---