Bifunctional Al2O3 Interlayer Leads to Enhanced ...

FULL PAPER Perovskite Solar Cells

www.solar-rrl.com

Bifunctional Al2O3 Interlayer Leads to Enhanced Open-Circuit Voltage for Hole-Conductor-Free Carbon-Based Perovskite Solar Cells Yuli Xiong, Xiaotong Zhu, Anyi Mei, Fei Qin, Shuang Liu, Shujing Zhang, Youyu Jiang, Yinhua Zhou, and Hongwei Han* Spiro-OMeTAD or PTAA and gold or silver are the most widely utilized HTM and CE, respectively, which represent a large portion of the cost of the PSCs. So, great efforts have been focused on the possibility to make HTM-free devices with carbon materials as the CE so far.[9–16] The most famous carbon-based PSC (C-PSC) was developed by Han’s group,[10,11,17] which has a typical architecture based on triple mesoporous layers of TiO2 (ETM), ZrO2 (Spacer layer), and C (CE). However, the HTM-free devices systematically showed a lower open circuit voltage (Voc) compared with the voltage achieved in similar devices with HTM. In PSCs, Voc is closely related to the recombination at the interfaces where the photo-induced electrons recombine with holes.[18] Such issue is more serious in HTM-free C-PSCs due to the direct contact of carbon and TiO2,[19,20] To avoid the contact, a thick spacer layer (2–3 μm) is inserted between CE and ETM, although the thick spacer layer is detrimental for charge transport. Thus, developing a novel spacer structure is meaningful for the improvement of Voc in C-PSCs. On the other hand, it has been demonstrated that change the electrical properties of ETMs by modifying ETM surface with insulating materials, such as Al2O3,[21,22] La2O3,[23] and MgO,[24] has successfully suppressed the charge recombination and improved the Voc of devices. Considering the two aspects above, combination of modification of ETM’s surface and spacer layer through architecture design is proposed to improve the Voc of HTM-free C-PSCs for the first time in this work. In this study, an Al2O3 layer is inserted between ETM and spacer layer by a facile process, which could modify the ETM’s surface and spacer layer at one time, while this generally should be carried out by two separate steps. Briefly, we deposited Al thin film on TiO2 (P25) film with organic species (ethyl cellulose) by vacuum evaporation. Then, during the further calcination in device’s fabrication process, the Al film was oxidized to Al2O3 and became partially porous Al2O3 layer, which made an ultrathin Al2O3 layer formed on the surface of mesopore channels inside TiO2 layer and a thin Al2O3 capping layer on P25 film. Synergetic effects of this Al2O3 interlayer in carbon-based HTM-free mesoscopic PSCs have been investigated in this work.

Hole-conductor-free carbon-based perovskite solar cell (C-PSC) is respected for their low cost and super stability. However, its absolute efficiency is hampered by the relatively low open-circuit voltage (Voc) due to the higher recombination losses than device with hole conductors. Herein, we develop a novel architecture to improve the Voc through simple incorporation of an Al2O3 interlayer. We find that this Al2O3 interlayer not only serves as an excellent insulating layer to separate cathode and anode but also modifies the interface between the electron transport material (ETM) and perovskite, and thus effectively retards the recombination at ETM/perovskite and perovskite/counter electrode interfaces simultaneously. Significantly, the average Voc of hole-conductor-free C-PSC upon the adoption of Al2O3 interlayer is increased from 836 to 942 mV. This novel architecture design combine the surface modification and spacer tuning technologies through a facile step, which makes an important step toward obtaining high efficient hole-conductor-free C-PSC.

1. Introduction Perovskite solar cells (PSC) have triggered worldwide intense research due to the ease of fabrication and excellent physical properties of perovskite materials, including high absorption coefficient, high mobility, long balanced carrier diffusion length, and low exciton binding energy.[1–7] Therefore, various optimization methods were attempted and the power conversion efficiency (PCE) has been boosted from 3.8% in 2009 to a certified 22.7% in 2017.[8] A typical PSC is composed of electron transport material (ETM), perovskite material, hole transporting material (HTM), and counter electrode (CE). Expensive Dr. Y. Xiong, X. Zhu, A. Mei, F. Qin, S. Liu, S. Zhang, Dr. Y. Jiang, Prof. Y. Zhou, Prof. H. Han Michael Grätzel Center for Mesoscopic Solar Cells Wuhan National Laboratory for Optoelectronics School of Optical and Electronic Information Huazhong University of Science and Technology Wuhan 430074, Hubei, P. R. China E-mail: [email protected] The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/solr.201800002.

DOI: 10.1002/solr.201800002

Sol. RRL 2018, 2, 1800002

1800002 (1 of 7)

© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com

www.solar-rrl.com

schematic diagram of the working principle of the holeconductor-free C-PSC with Al2O3 interlayer. Due to the alignment of energy levels of the device components, the electron transfers from perovskite (3.9 eV) to TiO2 (4.0 eV) and holes collects from perovskite (5.4 eV) to carbon layer (5.0 eV). Both the Al2O3 capping layer and the ZrO2 film serve as spacer layer to separate the ETM and carbon CE, which is an interface where electron–hole recombination can take place. The ultrathin Al2O3 inside TiO2(Al2O3) film modified the surface of ETM/perovskite, which would not affect the forward electron transport, but could prevent the electrons in TiO2 from recombining with holes left in perovskite layer. So, the additional Al2O3 interlayer suppressed the recombination process at interfaces from two aspects, which is expected to greatly improve the open-circuit voltage of carbon-based HTM-free 2. Results and Discussion C-PSCs. From the SEM cross-sectional image of TiO2/Al film Scheme 1A displays the formation processes of bifunctional Al2O3 interlayer. After the screen-printing of TiO2 (P25) films (Figure 1A), the TiO2 film underneath is not mesoporous due with organic species (ethyl cellulose), Al thin film was deposited to the existence of ethyle cellulose. The Al film is very flat and by vacuum evaporation. Then, during the further calcination in compact with a thickness of around 50 nm, laying only on the device’s fabrication process, the Al became Al2O3 and part of surface of TiO2 film, which is also suggested by the EDS Al2O3 fell inside the pores of TiO2 film, so the Al2O3 film became mapping of Al element. After calcination (Figure 1B), the TiO2 partially porous and an ultrathin Al2O3 layer formed on the film became mesoporous TiO2(Al2O3) film, and the Al2O3 surface of mesopore channels in TiO2 layer. The surfacecapping layer is mesoporous with a thickness of around 60 nm. modified TiO2 is defined as TiO2(Al2O3). After the infiltration of Although the Al2O3 film is mesoporous, it has bigger particle perovskite solution, the Al2O3 capping layer serves as an size with more flat and compact surface compared to the normal excellent insulating layer and the Al2O3 inside the TiO2(Al2O3) mesoporous ZrO2 with aggregations and cracks, as shown in Figure S1, Supporting Information, which plays a critical role in film could hinder the direct contact between TiO2 and the properties of spacer layer to further affect the device’s perovskite. Noticeably, the modification of the spacer layer performance.[25] The Al elemental mapping of TiO2/Al2O3 film and ETM’s surface through the bifunctional Al2O3 interlayer was carried out at one time, while this generally should be carried out confirms the Al element exists on the whole cross section of TiO2 by two separate steps. film, as the Al2O3 gets inside the ETM film through its mesopore The whole schematic structure of C-PSC with Al2O3 interlayer channels. This suggests the Al2O3 interlayer has dual functions, is shown in Scheme 1B. Mesoporous TiO2(Al2O3), ZrO2 and which forms not only a compact mesoporous capping layer on TiO2(Al2O3) film, but also an ultrathin film modified the surface carbon films are screen-printed on a Glass/FTO/C-TiO2 of TiO2. Figure 1C displays the SEM surface images of TiO2/ substrate layer by layer. Al2O3 bifunctional interlayer is inserted between mesoporous TiO2(Al2O3) and ZrO2 films. Then, Al2O3 film, it is clearly observed that the morphology of TiO2/ perovskite is filled into the mesoscopic layers directly with a Al2O3 film is absolutely different from TiO2 film (Figure S1, simple solution-processed method. Scheme 1C gives a simple Supporting Information), indicating that the Al2O3 capping layer is fully covered on TiO2 film. And the corresponding EDS mapping of Al and Ti elements further confirms the existence of Al2O3 film on the top of TiO2. In Figure 1D, compared to TiO2/Al2O3 film, the absolutely different morphology of TiO2(Al2O3) film suggests the totally removal of Al2O3 capping layer, while the EDS mapping of Al element approves that the Al2O3 gets inside the TiO2 film and decorates its surface. In Figure 2A, the existence of Al2O3 in TiO2/ Al2O3 and TiO2(Al2O3) film could be confirmed by the XPS spectra of Al 2p. Additionally, the signal of Ti 2p spectra in Figure 2B cannot be detected in TiO2/Al2O3 film, indicating the TiO2 film are fully covered by Al2O3 spacer layer. For TiO2(Al2O3) film, the Ti 2p peaks were shifted to a lower binding energy, indicating an electronic interaction between the TiO2 and Al2O3, and the modification is limited [26,27] Ultraviolet photoScheme 1. A) Formation process of bifunctional Al2O3 interlayer; schematic structure to the very surface of TiO2. electron spectroscopy (UPS) was further carried out (B) and energy band diagram (C) of hole-conductor-free C-PSCs with Al2O3 interlayer. We found that the presence of Al2O3 interlayer can greatly affect the Voc of cells, as the Al2O3 interlayer not only serves as an excellent insulating layer but also modifies the interface between ETM and perovskite, which further inhibited the recombination at ETM/perovskite and perovskite/CE interfaces at one time. As a result, with the adoption of this bifunctional Al2O3 interlayer, the average PCE of carbon-based HTM-free mesoscopic PSCs was increased from 11.46 to 13.42%, where the Voc raised from 836 to 942 mV particularly, and a highest PCE of 14.26% was realized. This suggests a convenient and effective method to obtain HTM-free C-PSCs with simple device structure and high cost-efficiency.

Sol. RRL 2018, 2, 1800002

1800002 (2 of 7)



www.solar-rrl.com

Figure 1. SEM cross-sectional images of (A) TiO2/Al and (B) TiO2/Al2O3 film and surface images of (C) TiO2/Al2O3 and (D) TiO2(Al2O3) film with corresponding EDS mapping of Al and Ti elements.

(Figure 2C) and displayed that the energy band level of TiO2(Al2O3) film was shifted compared to pristine TiO2, confirming the surface decoration of TiO2 by Al2O3. These are in accordance with SEM and EDS mapping results. From the XRD spectra of TiO2/Al2O3/ZrO2/Pero and TiO2/ ZrO2/Pero films (Figure 3D), the patterns of TiO2/Al2O3/ZrO2/ Pero and TiO2/ZrO2/Pero films display no differences, indicating that the additional Al2O3 interlayer did not affect the crystallization of perovskite inside. However, the charge transport character at ETM/perovskite was influenced by the surface modification of Al2O3, as suggested by time-resolved and steady-state photoluminescence (PL) results. For TiO2/ Pero, TiO2(Al2O3)/Pero films in Figure 2E, the excitons in perovskite transfers to ETM quicker in TiO2(Al2O3)/Pero film (3.58 ns) than that in TiO2/Pero films (4.64 ns). Furthermore, as illustrated in Table S1, Supporting Information, the fitted results suggest that the PL decay dynamics of TiO2/Pero and TiO2(Al2O3)/Pero samples consist of fast and slow decay components. The lifetime of the fast decay process was used to

identify the quenching effect of free carriers in the perovskite grains through transport to the ETM interface, whereas the slow decay process represents the results of radiative decay.[23,28] The time-resolved PL decay in which the fast decay lifetime was 1.97 ns for the TiO2/Pero film, which is reduced to 1.44 ns after Al2O3 interface modification on TiO2, indicating that charge transfer between perovksite layer and TiO2(Al2O3) is improved owing to the surface modification. These were also approved by the enhanced steady state PL for the samples (Figure 3F). The PL intensity of the TiO2(Al2O3)/Pero film is reduced compared with that of the TiO2/Pero film, indicating that the charge transfer effectively occurred before the carrier recombination at the interface and eventually improved the electron extraction rate from the perovskite absorber layer. The results of time-resolved and steady-state PL also supported the enhancement in PCE for the device with Al2O3 interlayer compared with that without the Al2O3 interlayer. The insulating property of spacer film was tested according to our previous report.[25] The insulating resistance (Table S2,

Figure 2. XPS spectra of Al 2p (A) and Ti 2p (B) and UPS spectra (C) for TiO2, TiO2/Al2O3, and TiO2(Al2O3) films; (D) XRD spectra of TiO2/Al2O3/ZrO2/ Pero and TiO2/ZrO2/Pero films; (E) time-resolved PL decays and (F) steady-state PL spectra of TiO2/Pero and TiO2(Al2O3)/Pero films.

Sol. RRL 2018, 2, 1800002

1800002 (3 of 7)



www.solar-rrl.com

60 nm Al2O3 interlayer contribute equally to the Voc factor of the HTM-free C-PSCs. This is because Al2O3 spacer layer is much more compact with bigger particles than ZrO2 and the interface of TiO2 was decorated by Al2O3 in C-PSCs with Al2O3 interlayer. Considering the huge thickness differences between 1 μm ZrO2 and 60 nm Al2O3 interlayer, the introduction of Al2O3 interlayer is proved to be an effective method for the improvement of Voc in HTM-free C-PSCs. The cross-sectional images of a typical C-PSC with Al2O3 interlayer before (top image) and after (bottom image) perovskite infiltration are displayed in Figure 3B. It is easily noticed that the Al2O3 interlayer is much more compact compared to mesoporous ZrO2 layers, which is good for the devices’ performance.[25] The top image also shows that both of the mesoporous TiO2 and ZrO2 layers have a thickness of 1 mm and a thin Al2O3 interlayer with a thickness of 60 nm is inserted between them. After the filling of perovskite, the Figure 3. A) Voc as a function of C-PSCs with various space layers; (B) cross-sectional perovskite penetrated into the pores of mesoporous image of the typical C-PSC with Al2O3 interlayer before and after perovskite infiltration; TiO2/Al2O3/ZrO2 very well, which created a homo(C) representative J–V curves of C-PSCs with various spacer layers. geneous compact film of both components while each material is very hard to be distinguished. J–V curves of C-PSCs based on various architecSupporting Information) reflects the insulating ability of tures were taken, as shown in Figure 3C. Without spacer layer, the spacer film. After the introduction of bifunctional Al2O3 C-PSC (TiO2/C) shows the lowest Voc due to the strong interlayer, the average resistance increased from 3.42 to recombination at perovskite/C interface, which reveals the critical 519.07 kΩ, indicating that Al2O3/ZrO2 spacer can prevent role of spacer in hole-conductor-free C-PSCs.[30] After the adoption carbon counter electrode from contacting with ETM more of 1 μm ZrO2 spacer, the Voc increased from 600 to 835 mV, due to effectively than ZrO2 spacer. the retarded recombination at perovskite/CE interface. SurprisSome preliminary photovoltaic experiments were conducted ingly, for device with 10 nm Al2O3 interlayer (60nm real thickness), to evaluate the performance of devices by varying the thickness it shows the same Voc as 1 μm ZrO2 spacer. The Voc increased from of Al2O3 and ZrO2 films. The thickness of Al2O3 interlayer 600 to 835 mV, indicating an effective reduced recombination at (marked as 0, 5, 10, 15, 20 nm) is the primary setting thickness of ETM/perovskite and perovskite/CE interfaces for bifunctional equipment during the vacuum evaporation, not stands for the Al2O3 interlayer in C-PSCs. The highest Voc of 950 mV is obtained real thickness of Al2O3 interlayer. With the introduction of 10 nm for C-PSC with 10 nm Al2O3 interlayer and 1 μm ZrO2, as it has the Al2O3 interlayer (Figure 3A), the Voc increases along with the most effective spacer layer with modified TiO2 surface. increase of ZrO2 thickness from 0 to 1 μm, when even thicker As shown in Figure 4A, the electron recombination from ZrO2 layer was applied, the Voc decreased a little. This is because IMVS is slower in the device compared with Al2O3 interlayer the increase of ZrO2 thickness would improve the insulating than that without Al2O3 interlayer. This is in agreement with the ability to avoid the quick recombination at perovskite/CE improved photovoltaic performance of devices by introduction of interface, however, the transport distance of electron and holes bifunctional Al2O3 interlayer. Figure 4B shows the Nyquist plots will increase which adds the possibility of recombination and recorded for device with and without Al2O3 in dark. From our further decrease the Voc, if the ZrO2 thickness is too large.[29] So, previous report, the RC response in the high-frequency region is 1 μm was considered as the optimum thickness of ZrO2. Then, assigned to the electron transfer process in device, whereas the we investigated the effects of Al2O3 thickness with 1 μm ZrO2 on low frequency region represents the recombination process in the Voc of HTM-free C-PSCs. With the increase of Al2O3 device.[10] The finding is that the device with Al2O3 has faster thickness, the Voc increased quickly then decreased sharply with electron transfer and slower recombination, which is in the optimum thickness of 10 nm. This is because the insulating accordance with other testing results. ability boosted with the increase of Al2O3 thickness. However, Figure 4C and Tables S3 and S4, Supporting Information thicker Al2O3 make the device be more difficult infiltrated. As summarize the statistical data concerning the numerical shown in Figure 3B and Figure S2, Supporting Information, the distribution of the key photovoltaic parameters for devices with perovskite were well infiltrated in device with 5 and 10 nm Al2O3 and without Al2O3 interlayer. After the introduction of Al2O3 interlayer, while the lower part of TiO2 layer in device with 15 and interlayer, the average PCE was improved from 11.46 to 13.42%, 20 nm Al2O3 is not well infiltrated. Surprisingly, the average Voc specifically, the Voc was increased from 836 to 942 mV. This is of device with 1 μm ZrO2 and 10 nm Al2O3 (around 60 nm in real because of reduced recombination at TiO2/perovskite and thickness) are quite similar, which suggests that 1 μm ZrO2 and perovksite/C interfaces. It should be noticed that only a small

Sol. RRL 2018, 2, 1800002

1800002 (4 of 7)



www.solar-rrl.com

improvement in HTM-free C-PSCs. The IPCE spectra and corresponding integrated photocurrents of the cell with Al2O3 interlayer is displayed in Figure 4E. The integrated photocurrents is 18.4 mA cm2, which agrees closely with the photocurrent density of 19.1 mA cm2 measured at the beginning of testing, which rose to 23.3 mA cm2 after 3 min of light soaking, although the theory has not been fully understood yet.[17]

3. Conclusion

Figure 4. A) Electron recombination time as a function of the Voc, obtained through IMVS; (B) EIS spectra of C-PSC with and without Al2O3 interlayer in dark; (C) histograms of the photovoltaic parameters for 20 separate C-PSCs with and without Al2O3 interlayer; (D) maximal steady-state photocurrent output at the maximum power point for the champion cell at 700 mV; (E) IPCE spectra with corresponding integrated current of champion cell.

deviation were observed for these devices, which indicates the perfect reproductivity of these devices. The representative J–V curves of C-PSCs with Al2O3 interlayer recorded in reverse and forward directions are displayed in Figure S3, Supporting Information. It has a Voc of 950 mV, Jsc of 22.89 mA cm2, FF of 62%, and PCE of 13.45% in reverse direction, while those are 960 mV, 21.55 mA cm2, 64 and 13.22% in forward direction. It can be found that the C-PSCs with Al2O3 interlayer did not show obvious hysteresis. As shown in Figure 4D, the champion cell with Al2O3 interlayer displays a Voc of 930 mV, Jsc of 23.3 mA cm2, FF of 66%, yielding a PCE of 14.26%. Although higher efficiencies of HTM-free C-PSCs have been obtained in our or others’ group by means of the adjustion of materials,[1,31–36] it is among the highest HTM-free CPSCs efficiencies based on 5-AVA modified perovskite reported to date.[17,25,37–39] Table S5, Supporting Information summarizes the typical device performance of the HTM-free C-PSCs based on ZrO2 or Al2O3 spacer layer. By the adoption of either ZrO2 or Al2O3 as spacer layer, C-PSCs could show good device performance,[15,17,33,35,38–42] although Fan et al. suggest the ZrO2 spacer layer are beneficial to the infiltration and crystallization of perovskite compared with Al2O3.[43] In future work, more efficient HTM-free C-PSCs could be made by combination of material tuning and this novel architecture. So, it is a general way for the

Sol. RRL 2018, 2, 1800002

In this work, we developed a novel architecture to improve the Voc through the incorporation of bifunctional Al2O3 layer between ETM and spacer layer by a facile process. From the results of SEM, EDX mapping, XPS, and UPS, it is confirmed that the Al2O3 layer modified the ETM’s surface and spacer layer. Importantly, the presence of Al2O3 interlay can greatly inhibit the recombination at ETM/perovskite and perovskite/CE interfaces at one time from the results of time-resolved and steadystate PL, IMVS, and EIS techniques. As a result, the Voc of carbon-based HTM-free mesoscopic PSCs based on this novel architecture raised from 836 to 942 mV and the PCE was increased from 11.46 to 13.42% with a highest PCE of 14.26%. This suggests a general method for the Voc improvement and a convenient and effective method to obtain HTM-free C-PSCs with high cost-efficiency.

4. Experimental Section

Device Fabrication: FTO-coated glasses were firstly etched with a laser to obtain the required electrode pattern. The sheets were then sequential ultrasonically cleaned for 20 min in dilute detergent, deionized water, and ethanol, followed by drying with hot air. A TiO2 compact layer was then deposited on the glass by spray pyrolysis deposition at 450 C with using di-isopropoxytitanium bis(acetylacetonate) (Sigma– Aldrich) diluted in ethanol. The TiO2, ZrO2, and C paste was fabricated according to previously published methods.[17] The TiO2 monocrystalline layer (P25) was deposited on top of the TiO2 compact layer by screen printing and then heated at 75 C for 30 min. Then, Al film with different thickness was deposited by vacuum evaporation and sintered at 500 C for 40 min. Followed a ZrO2 spacer layer and carbon layer were screen-printed and then sintered at 400 C for 40 min. After cooling down to the room temperature, the 35 wt.% (5-AVA)x(MA)1xPbI3 perovskite precursor solution (0.5475 g MAI, 1.6158 g PbI2, and 0.02994 g 5-AVAI were dissolved in 3.6 ml γ-butyrolactone and then stirred at 60 C overnight) was infiltrated by drop casting via the top of the carbon counter electrode. After drying at 50 C more than 1 h, the mesoscopic solar cells containing perovskite were obtained. Characterization: All the J–V curves in this study were recorded using a Keithley 2400 source meter unit. The device photocurrent was measured under AM1.5 illumination condition at an intensity of 100 mW cm2. The area of aperture is 0.107 cm2. The illumination intensity of the light source was accurately calibrated with a standard Si photodiode detector equipped with a KG-5 filter, which can be traced back to the standard cell of the National Renewable Energy Laboratory (NREL). The EQE spectra performed here were obtained from an IPCE setup consisting of a Xenon lamp (Oriel, 450 W) as the light source, a monochromator, a

1800002 (5 of 7)



www.solar-rrl.com

chopper with a frequency of 100 Hz, a lock-in amplifier (SR830, Stanford Research Corp), and a Si-based diode (J115711-1-Si detector) for calibration. The electron recombination time were measured by IMVS (Zahner, Zennium) at different bias light intensities. The film thickness and the SEM pictures of the devices were obtained by a FESEM (Hitachi Model S-4800 field-emission scanning electron microscopy). The XRD spectra were obtained by a Bruker D8 Advance X-ray diffractometer with Cu Kα radiation (λ ¼ 1.5418 Å ). Time-resolved photoluminescence (PL) decay transients were measured at 765 nm using excitation with a 478 nm light pulse from the Horiba Scientific DeltaPro. The steady-state PL was measured on LabRAM HR800 (Horiba Jobin Yvon) with the excitation wavelength of 532 nm. The XPS measurements were performed on an AXIS-ULTRA DLD-600W instrument and its radiation was generated by a He-gas discharge lamp (He 1α at 21.22 eV).

Supporting Information Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgments Y. Xiong and X. Zhu contributed equally to this work. The authors acknowledge financial support from the National Natural Science Foundation of China (Grant Nos. 91433203, 61474049, 51502141, and 21702069), the Science and Technology Department of Hubei Province (No. 2017AAA190), the 111 Project (No. B07038), the China Postdoctoral Science Foundation (2016M602292), Special financial aid to post-doctor research fellow (2017T100548). We thank the Analytical and Testing Center of Huazhong University of Science and Technology (HUST) for performing various characterization and measurements.

Conflict of Interest The authors declare no conflict of interest.

Keywords bifunctional Al2O3, hole-conductor-free, insulating, perovskite solar cells, surface modification Received: January 5, 2018 Revised: February 1, 2018 Published online: February 22, 2018

[1] H. Chen, S. Yang, Adv. Mater. 2017, 29, 1603994. [2] J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin, M. Grätzel, Nature 2013, 499, 316. [3] H. Zhou, Q. Chen, G. Li, S. Luo, T.-b. Song, H.-S. Duan, Z. Hong, J. You, Y. Liu, Y. Yang, Science 2014, 345, 542. [4] A. Polman, M. Knight, E. C. Garnett, B. Ehrler, W. C. Sinke, Science 2016, 352, 307. [5] M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami, H. J. Snaith, Science 2012, 338, 643. [6] 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, N.-G. Park, Sci. Rep. 2012, 2, 591. [7] W. A. Laban, L. Etgar, Energy Environ. Sci. 2013, 6, 3249. [8] http://www.nrel.gov/ncpv/images/efficiency_chart.jpg.

Sol. RRL 2018, 2, 1800002

[9] J. Chen, Y. Rong, A. Mei, Y. Xiong, T. Liu, Y. Sheng, P. Jiang, L. Hong, Y. Guan, X. Zhu, X. Hou, M. Duan, J. Zhao, X. Li, H. Han, Adv. Energy Mater. 2016, 6, 1502009. [10] Y. Rong, Z. Ku, A. Mei, T. Liu, M. Xu, S. Ko, X. Li, H. Han, J. Phys. Chem. Lett. 2014, 5, 2160. [11] J. Chen, Y. Xiong, Y. Rong, A. Mei, Y. Sheng, P. Jiang, Y. Hu, X. Li, H. Han, Nano Energy 2016, 27, 130. [12] Z. Ku, Y. Rong, M. Xu, T. Liu, H. Han, Sci. Rep. 2013, 3, 3132. [13] H. Chen, Z. Wei, H. He, X. Zheng, K. S. Wong, S. Yang, Adv. Energy Mater. 2016, 6, 1502087. [14] Z. Wei, K. Yan, H. Chen, Y. Yi, T. Zhang, X. Long, J. Li, L. Zhang, J. Wang, S. Yang, Energy Environ. Sci. 2014, 7, 3326. [15] K. Cao, Z. Zuo, J. Cui, Y. Shen, T. Moehl, S. M. Zakeeruddin, M. Grätzel, M. Wang, Nano Energy 2015, 17, 171. [16] X. Xu, Z. Liu, Z. Zuo, M. Zhang, Z. Zhao, Y. Shen, H. Zhou, Q. Chen, Y. Yang, M. Wang, Nano Lett. 2015, 15, 2402. [17] A. Mei, X. Li, L. Liu, Z. Ku, T. Liu, Y. Rong, M. Xu, M. Hu, J. Chen, Y. Yang, M. Grätzel, H. Han, Science 2014, 345, 295. [18] N. K. Elumalai, A. Uddin, Energy Environ. Sci. 2016, 9, 391. [19] S. Guarnera, A. Abate, W. Zhang, J. M. Foster, G. Richardson, A. Petrozza, H. J. Snaith, J. Phys. Chem. Lett. 2015, 6, 432. [20] Y. Yang, H. Chen, X. Zheng, X. Meng, T. Zhang, C. Hu, Y. Bai, S. Xiao, S. Yang, Nano Energy 2017, 42, 322. [21] Y. H. Lee, J. Luo, M.-K. Son, P. Gao, K. T. Cho, J. Seo, S. M. Zakeeruddin, M. Grätzel, M. K. Nazeeruddin, Adv. Mater. 2016, 28, 3966. [22] J. M. Marin-Beloqui, L. Lanzetta, E. Palomares, Chem. Mater. 2016, 28, 207. [23] S. Shaikh, H.-C. Kwon, W. Yang, H. Hwang, H. Lee, E. Lee, S. I. Ma, J. Moon, J. Mater. Chem. A 2016, 4, 15478. [24] G. S. Han, H. S. Chung, B. J. Kim, D. H. Kim, J. W. Lee, B. S. Swain, K. Mahmood, J. S. Yoo, N.-G. Park, J. H. Lee, H. S. Jung, J. Mater. Chem. A 2015, 3, 9160. [25] T. Liu, Y. Rong, Y. Xiong, A. Mei, Y. Hu, Y. Sheng, P. Jiang, X. Hou, M. Duan, Y. Guan, L. Hong, H. Han, RSC Adv. 2017, 7, 10118. [26] D. Yang, X. Zhou, R. Yang, Z. Yang, W. Yu, X. Wang, C. Li, S. Liu, R. P. H. Chang, Energy Environ. Sci. 2016, 3071. [27] H. P. Dong, Y. Li, S. F. Wang, W. Z. Li, N. Li, X. D. Guo, L. D. Wang, J. Mater. Chem. A 2015, 3, 9999. [28] X. Zhao, H. Shen, Y. Zhang, X. Li, X. Zhao, M. Tai, J. Li, J. Li, X. Li, H. Lin, ACS Appl. Mater. Interfaces 2016, 8, 7826. [29] T. Liu, L. Liu, M. Hu, Y. Yang, L. Zhang, A. Mei, H. Han, J. Power Sources 2015, 293, 533. [30] Y. Hu, S. Si, A. Mei, Y. Rong, H. Liu, X. Li, H. Han, Solar RRL 2017, 1, 1600019. [31] X. Zheng, H. Chen, Q. Li, Y. Yang, Z. Wei, Y. Bai, Y. Qiu, D. Zhou, K. S. Wong, S. Yang, Nano Lett. 2017, 17, 2496. [32] X. Hou, Y. Hu, H. Liu, A. Mei, X. Li, M. Duan, G. Zhang, Y. Rong, H. Han, J. Mater. Chem. A 2017, 5, 73. [33] Y. Rong, X. Hou, Y. Hu, A. Mei, L. Liu, P. Wang, H. Han, Nature Comm. 2017, 8, 14555. [34] M. Duan, Y. Hu, A. Mei, Y. Rong, H. Han, Mater. Today Energy 2017, https://doi.org/10.1016/j.mtener.2017.09.016. [35] L. Hong, Y. Hu, A. Mei, Y. Sheng, P. Jiang, C. Tian, Y. Rong, H. Han, Adv. Funct. Mater. 2017, 27, 1703060. [36] Y. Sheng, Y. Hu, A. Mei, P. Jiang, X. Hou, M. Duan, L. Hong, Y. Guan, Y. Rong, Y. Xiong, H. Han, J. Mater. Chem. A 2016, 4, 16731. [37] M. Duan, C. Tian, Y. Hu, A. Mei, Y. Rong, Y. Xiong, M. Xu, Y. Sheng, P. Jiang, X. Hou, X. Zhu, F. Qin, H. Han, ACS Appl. Mater. Interfaces 2017, 9, 31721. [38] M. Duan, Y. Rong, A. Mei, Y. Hu, Y. Sheng, Y. Guan, H. Han, Carbon 2017, 120, 71. [39] P. Jiang, T. W. Jones, N. W. Duffy, K. F. Anderson, R. Bennett, M. Grigore, P. Marvig, Y. Xiong, T. Liu, Y. Sheng, L. Hong, X. Hou, M. Duan, Y. Hu, Y. Rong, G. J. Wilson, H. Han, Carbon 2018, 129, 830.

1800002 (6 of 7)



www.solar-rrl.com

[40] H. Li, K. Cao, J. Cui, S. Liu, X. Qiao, Y. Shen, M. Wang, Nanoscale 2016, 8, 6379. [41] C.-M. Tsai, G.-W. Wu, S. Narra, H.-M. Chang, N. Mohanta, H.-P. Wu, C.-L. Wang, E.W. -G. Diau, J. Mater. Chem. A 2017, 5, 739.

Sol. RRL 2018, 2, 1800002

[42] C.-Y. Chan, Y. Wang, G.-W. Wu, E. Wei-Guang Diau, J. Mater. Chem. A 2016, 4, 3872. [43] Z. Meng, D. Guo, J. Yu, K. Fan, Appl. Surf. Sci. 2018, 430, 632.

1800002 (7 of 7)
