PCCP COMMUNICATION. Mesoporous SnO 2 single crystals as an effective electron collector for perovskite solar cells

Transcription

1 PCCP COMMUNICATION View Article Online View Journal View Issue Cite this: Phys. Chem. Chem. Phys., 2015, 17, Received 16th March 2015, Accepted 8th June 2015 Mesoporous SnO 2 single crystals as an effective electron collector for perovskite solar cells Zonglong Zhu, ab Xiaoli Zheng, b Yang Bai, b Teng Zhang, ab Zilong Wang, b Shuang Xiao ab and Shihe Yang* ab DOI: /c5cp01534k Mesoporous single crystals are prized for their fast electron transport and high surface area. Here we report the first synthesis of mesoporous SnO 2 single crystals (SnO 2 MSCs) by a simple silicatemplated hydrothermal method, and its application in solutionprocessed perovskite solar cells (PSCs). A relatively low efficiency (3.76%) was obtained due to the strong charge recombination at the SnO 2 /perovskite interface. However, by coating a thin TiO 2 barrier layer on SnO 2 via TiCl 4 treatment, we were able to achieve an 8.54% power conversion efficiency (PCE). A dynamics study using impedance spectroscopy revealed a much lower transport resistance for the SnO 2 MSC-based solar cells than for the TiO 2 nanocrystal PSCs, but a stronger recombination. Significantly, the thin TiO 2 coating layer on SnO 2 considerably reduced the recombination while largely maintaining the superior electron-transport properties. Perovskite solar cells (PSCs) based on organic inorganic hybrid organolead halide absorbers have developed dramatically and have become the most competitive candidate for use as thirdgeneration photovoltaics. 1 In recent years, PSCs have achieved over 15% power conversion efficiency (PCE), both in nanostructured sensitized solar cells and thin film solar cells. 2,3 The general formula of perovskite is CH 3 NH 3 PbX 3 (where X = halogen) and its structures generally have cuboctahedral and octahedral geometries. 4 6 The advantages of these halide perovskites have already been shown by their interesting optical, electroluminescence, magnetic and other properties. 7 Therefore, solar cells made from such halide perovskite absorbers hold the promise of advancing the field by being at the same time highly efficient and inexpensive. 8 a Nano Science and Technology Program, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong. chsyang@ust.hk b Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong Electronic supplementary information (ESI) available: Experimental details, SEM images, TEM images, element mapping of TiO 2 SnO 2 MSC, the overall distributions of device performance parameters, and the integrated photocurrent densities of the IPCE spectra. See DOI: /c5cp01534k These authors contributed equally. Electron collectors, i.e., acceptors, are widely used and necessary in the structure of PSCs. Mesoporous PSCs as well as thin film p i nheterojunctionsolarcellsneedathindensesemiconductor oxide layer (mostly TiO 2 )asanelectronacceptortofacilitatethe injection of electrons into the conducting substrate, while inverted PSCs use PCBM as the top layer to attract electrons to the metal electrode Therefore, an efficient electron collector would promote separation of excitons and furthermore improve the PCE. Recently, TiO 2 mesoporous single crystals (MSCs) were used in PSCs as an electron collector due to their fast electron transport and high surface area. 13,14 The MSCs should benefit the performance of solar cells both by providing high electron mobility and few bulk defects and impurities. Compared with TiO 2,tin(IV)oxide(SnO 2 )MSCsappeartobeanevenbetterchoice for PSC applications in many respects. 15 First of all, SnO 2 has a higher band gap (at 3.8 ev), which results in lower photocatalytic activity and higher device stability. Secondly, the charge mobility of SnO 2 is almost two orders of magnitude higher than that of TiO 2, 16,17 and even the trap density is lower, and the conduction band edge exhibits a 300 mv positive shift. 18 Here, we for the first time synthesized mesoporous SnO 2 MSCs by a simple silica-templated hydrothermal method, and further explored their use as the electron collector in PSCs deposited by a two-step CH 3 NH 3 PbI 3 solution process. The strong recombination in the SnO 2 MSC-based PSCs was found to limit the efficiency to less than 4.0% (highest efficiency B3.76%) due to the low photovoltage and photocurrent. After treating the SnO 2 MSCs with TiCl 4 to make a thin TiO 2 layer, the cell efficiency was dramatically increased to as high as 8.54%, indeed much higher than the 7.20% record for TiO 2 MSC-based PSCs. 13 Through a dynamics study by impedance spectroscopy, we found that the SnO 2 MSC-based PSCs present less transport resistance than do TiO 2 nanocrystal-based PSCs, but charge recombination in the devices is stronger. Significantly, the device based on the TiO 2 -coated SnO 2 (TiO 2 -SnO 2 ) MSCs retained the transport properties of that based on the pristine SnO 2 MSCs but with much less recombination. The synthesis of SnO 2 MSCs is an extension of our previous work on TiO 2 MSCs. 19 The scanning electron microscopy (SEM) This journal is the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17,

2 View Article Online Communication Fig. 1 SEM image (A), TEM images (B, C), selected-area electron diffraction (SAED) pattern of C (D), HRTEM image (E), and XRD patterns (F) of the SnO 2 MSCs. PCCP by spin-coating them onto separate fluorine-doped tin oxide (FTO) substrates, and annealing under 400 1C to increase the connectivity between the SnO 2 MSCs. The TiO 2 SnO 2 MSC layer deposition was similar to the method of fabricating SnO 2 dyesensitized solar cells, where the SnO 2 MSC films were treated in 50 mm TiCl 4 at 70 1C for 30 min. Fig. S1 (ESI ) shows that the resulting TiO 2 thin layer formed with a thickness of nm and much rougher surfaces than did pristine SnO 2 MSC, due to the 10 nm of TiO 2 nanoparticles grown thereon. The elemental mapping analysis (Fig. S2 and S3, ESI ) confirmed that the TiO 2 nanoparticles were uniformly distributed on the SnO 2 MSC surfaces. In contrast to the TiO 2 MSC-based PSCs, the SnO 2 MSCs removetheneedforthetio 2 compact layer, thus simplifying the device fabrication procedure. The CH 3 NH 3 PbI 3 and spiro-ometad layers were successively deposited onto both the SnO 2 MSC and TiO 2 SnO 2 MSCs films (for details, see ESI experimental section). The energy diagram and a cross-section image of a typical TiO 2 SnO 2 MSC-based PSC are shown in Fig. 2A and B. Here, the total thickness of the device was observed to be about nm, including a 300 nm-thick SnO 2 MSC layer, nm-thick spiro-ometad layer and 50 nm-thick Au electrode. The SnO 2 /CH 3 NH 3 PbI 3 /hole-transport layer (HTL)/Au and TiO 2 SnO 2 /CH 3 NH 3 PbI 3 /HTL/Au devices were examined under a simulated air mass 1.5 global (AM1.5G) solar irradiation and the active area of the electrode was set at 7 mm 2. Their respective current voltage ( J V) curves are shown in Fig. 3A, related photovoltaic parameters are summarized in Table 1, the overall and transmission electron microscopy (TEM) images in Fig. 1A and B both show the well-defined and homogeneously dispersed mesoporous nanorod-shaped SnO 2 MSCs, with an average width of nm and length of 500 nm 1 mm. With the assistance of the silica template, the mesopores with a diameter of B50 nm were obtained after removing the template, as shown in Fig. 1C. The Laue electron diffraction pattern shown in Fig. 1D indicated the single-crystal nature of SnO 2 MSCs enclosed by dominant (110) facets. The high-resolution TEM (HRTEM) image (Fig. 1E) clearly showed the well-ordered SnO 2 single crystals, whose outer d spacing was measured to be 0.33 nm and match well with the interplanar spacing of the (110) plane. The powder X-ray diffraction (XRD) pattern (Fig. 1F) established the phase of the SnO 2 MSCs with a tetragonal structure and all of the conspicuous peaks matched well with the standard pattern in JCPDS No After coating with TiO 2, the MSCs still maintained the mesoporous nanorod structure but the surfaces were observed to be much rougher due to the coating of a nm-thick layer of TiO 2, as shown in Fig. S1 (ESI ). It is clearly seen from the elemental mapping (Fig. S2 and S3, ESI ) that the Ti was indeed distributed uniformly on the SnO 2 MSC surface. The deposited TiO 2 thin surface layer was used to modify the chemical and electrical properties of SnO 2 MSCs, with the aim of decreasing the recombination from SnO 2 MSC to perovskite and further improving the PV efficiency of photon harvesting. For photoanode fabrication, we deposited both a SnO 2 MSC layer and a TiO 2 SnO 2 MSC layer, each one about 300 nm thick, Fig. 2 (A) Energy level alignment diagram of perovskite photovoltaic devices based on SnO 2 MSCs. 18,20 Note that upon the treatment of the SnO 2 MSC film with TiCl 4, a thin TiO 2 layer was coated. (B) A cross-sectional view of a typical TiO 2 SnO 2 MSC-based perovskite solar cell Phys. Chem. Chem. Phys., 2015, 17, This journal is the Owner Societies 2015

3 View Article Online PCCP Fig. 3 Typical photocurrent density versus applied voltage (J V) curves (A) and the IPCE spectra (B) of the perovskite solar cells. (A) TiO 2 SnO 2 MSC film is denoted by the black line and the SnO 2 MSC film by the red dashed line. (B) TiO 2 SnO 2 MSC is denoted by black squares and SnO 2 MSC by red circles. Table 1 Performance parameters of the SnO 2 MSCs, TiO 2 SnO 2 MSCs based perovskite solar cells Devices V OC (V) J SC (ma cm 2 ) FF PCE (%) TiO 2 SnO 2 MSC Pristine SnO 2 MSC distributions of device performance parameters are shown in Fig. S4 and S5 (ESI ), and the average performances of TiO 2 SnO 2 MSCs and SnO 2 MSCs PSCs are tabulated in Table S1 (ESI ). The TiO 2 SnO 2 MSC-based PSCs attained up to 8.54% PCE, with a short-circuit photocurrent (J SC ) of ma cm 2, open-circuit voltage (V OC ) of V and fill factor (FF) of 0.621, whereas the pristine (i.e., without TiCl 4 treatment) SnO 2 MSC-based device only presented a 3.76% PCE with a J SC of ma cm 2, V OC of V and FF of The efficiency difference was mainly caused by the J SC and V OC. First, it can be seen that the open-circuit voltage of the TiO 2 SnO 2 MSC cells (V OC = 0.802) is higher than that of pristine SnO 2 MSC cells (V OC = 0.546). This V decrease could be explained by the conduction band edge of SnO 2 being 300 mv more positive than that of TiO 2. The V OC of TiO 2 SnO 2 MSC devices was improved greatly by the TiO 2 modification, which shifted the electronic band to a more negative value and decreased the interfacial recombination. Then the photocurrent was also enhanced by the TiO 2 treatment process: for the pristine SnO 2 MSC device, J SC was measured to be only ma cm 2, whereas the TiO 2 SnO 2 MSC device Communication exhibited a J SC of ma cm 2. The photocurrent difference could be explained by the strong recombination of the pristine SnO 2 MSC device, and the details are discussed below. The incident photon-to-electron conversion efficiency (IPCE) spectra in Fig. 3B show an onset of B800 nm, and appear to be dominated by CH 3 NH 3 PbI 3 absorptions. The IPCE results agree well with the photocurrent from the J V curves, and the value of integrated photocurrent from IPCE is similar to the J SC (Fig. S6, ESI ), validating the measurements. To check whether the notable PCE for our devices is reliable, we took the stabilized photocurrent at a photovoltage near the maximum power point of the optimal device and monitored it as a function of time. As shown in Fig. S7 (ESI ), we obtained a steady-state PCE of 8.4% for the device based on the TiO 2 SnO 2 MSCs and 3.5% for that based on the SnO 2 MSCs, thereby confirming the PCE values derived from the J V curves. For comparison, a TiO 2 nanocrystal-based device without the TiO 2 dense layer but treated with TiCl 4 was also prepared, and displayed an efficiency of 8.61%, which is quite similar to the efficiency of TiO 2 SnO 2 MSCs (Fig. S8, ESI ). Electrochemical impedance spectroscopy (EIS) was used to investigate the interfacial charge transfer at the SnO 2 /CH 3 NH 3 PbI 3 / spiro-ometad interfaces. 21 We measured the EIS spectra at an applied bias of V OC (0.546 V, V and V for SnO 2 MSCs, TiO 2 SnO 2 MSCs, TiO 2 nanocrystal based PSC, respectively) and a frequency range of 1 Hz to 1 MHz with an AC amplitude of 10 mv under illumination of simulated solar AM1.5 global light at 100 mw cm 2. Fig. 4 shows the Nyquist plots derived from the results of EIS spectra of SnO 2 MSC, TiO 2 SnO 2 MSC and TiO 2 nanocrystal PSC films. The related parameters are summarized in Table S2 (ESI ). A simplified circuit model reported previously was used to fit these Nyquist curves (see the inset of Fig. 4). 22 Here, two clear semicircles were observed, one in the high-frequency region and the other in the low-frequency region. The first arc at high frequencies corresponds to the charge-transport behavior of the whole device. Since all of the devices showed similar values of R 1 (see Table S2, ESI ) and since these devices were designed to share the same structure, the transport properties also appeared to be similar. The low-frequency arc or the main arc is ascribed to the Fig. 4 EIS Nyquist profiles of the perovskite solar cells. Inset: the equivalent circuit used for fitting the Nyquist plots. Legends: TiO 2 SnO 2 MSC (red squares), SnO 2 MSC (black circles), TiO 2 nanocrystals (blue triangles). This journal is the Owner Societies 2015 Phys. Chem. Chem. Phys., 2015, 17,

4 View Article Online Communication recombination resistance (R 2 ) together with the chemical capacitance (C 2 ). By using the above features, we fitted the Nyquist plots by the appropriate equivalent models in the inset of Fig. 4. This fitting clearly showed the order of the recombination resistance being R 2(SnO2 MSC) o R 2(TiO2 SnO 2 MSC) o R 2(TiO2 nanocrystals), meaning that SnO 2 MSCs showed stronger recombination than did the TiO 2 MNC film. The coated TiO 2 thin layer decreased the recombination rate, but to a level still higher than that of the TiO 2 MNC film. This result explains why treatment with TiCl 4 could cause a large increase in the V OC. Conclusions In summary, we have for the first time synthesized SnO 2 MSC nanorods by a simple silica-templated growth method. We have also demonstrated the application of these SnO 2 MSCs in perovskite solar cells. When treated with TiCl 4,theSnO 2 MSC-based PSCs attained a record 8.54% efficiency, which is higher than the efficiency values of all of the SnO 2 -based DSSCs as well as of the TiO 2 MSC-based PSCs, demonstrating their potential to rival the mesoporous TiO 2 PSCs. The EIS measurements further showed a very good electron-transport ability for the SnO 2 mesoporous single crystals, which makes these crystals a most promising electron collector for PSCs. Conflicts of interest The authors declare no competing financial interests. We thank Prof. He Yan for providing access to the glovebox facility. Acknowledgements This work was supported by the HK-ITF (ITS/004/14) and HK-RGC General Research Funds (HKUST ). Notes and references 1 N. G. Park, J. Phys. Chem. Lett., 2013, 4, M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Gratzel, Nature, 2013, 499, PCCP 4 H.-S. Kim, S. H. Im and N.-G. Park, J. Phys. Chem. C, 2014, 118, E. Mosconi, A. Amat, M. K. Nazeeruddin, M. Gratzel and F. De Angelis, J. Phys. Chem. C, 2013, 117, J. H. Qiu, Y. C. Qiu, K. Y. Yan, M. Zhong, C. Mu, H. Yan and S. H. Yang, Nanoscale, 2013, 5, H. J. Snaith, J. Phys. Chem. Lett., 2013, 4, M. A. Loi and J. C. Hummelen, Nat. Mater., 2013, 12, S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, O. Malinkiewicz, A. Yella, Y. H. Lee, G. M. Espallargas, M. Graetzel, M. K. Nazeeruddin and H. J. Bolink, Nat. Photonics, 2014, 8, Q. Chen, H. Zhou, Z. Hong, S. Luo, H.-S. Duan, H.-H. Wang, Y. Liu, G. Li and Y. Yang, J. Am. Chem. Soc., 2013, 136, Z. Zhu, J. Ma, Z. Wang, C. Mu, Z. Fan, L. Du, Y. Bai, L. Fan, H. Yan, D. L. Phillips and S. Yang, J. Am. Chem. Soc., 2014, 136, E. J. W. Crossland, N. Noel, V. Sivaram, T. Leijtens, J. A. Alexander-Webber and H. J. Snaith, Nature, 2013, 495, S. So and P. Schmuki, Angew. Chem., Int. Ed., 2013, 52, G.Sadoughi,V.Sivaram,R.Gunning,P.Docampo,I.Bruder, N. Pschirer, A. Irajizad and H. J. Snaith, Phys.Chem.Chem.Phys., 2013, 15, E. Yagi, R. R. Hasiguti and M. Aono, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, Z. M. Jarzebski and J. P. Marton, J. Electrochem. Soc., 1976, 123, A. N. M. Green, E. Palomares, S. A. Haque, J. M. Kroon and J. R. Durrant, J. Phys. Chem. B, 2005, 109, X. Zheng, Q. Kuang, K. Yan, Y. Qiu, J. Qiu and S. Yang, ACS Appl. Mater. Interfaces, 2013, 5, H. J. Snaith and C. Ducati, Nano Lett., 2010, 10, Q. Wang, J. E. Moser and M. Gratzel, J. Phys. Chem. B, 2005, 109, L. Dloczik, O. Ileperuma, I. Lauermann, L. M. Peter, E. A. Ponomarev, G. Redmond, N. J. Shaw and I. Uhlendorf, J. Phys. Chem. B, 1997, 101, Phys. Chem. Chem. Phys., 2015, 17, This journal is the Owner Societies 2015

5 Electronic Supplementary Material (ESI) for Physical Chemistry Chemical Physics. This journal is the Owner Societies 2015 Mesoporous SnO 2 Single Crystals as an Effective Electron Collector for Perovskite Solar Cells Zonglong Zhu 1, Xiaoli Zheng 1, Yang Bai, Teng Zhang, Zilong Wang, Shuang Xiao, Shihe Yang * Nano Science and Technology Program, Department of Chemistry, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

6 Experimental section Preparation of SiO 2 colloid. Colloid silica nanospheres (diameter: 50 nm) were synthesized by mixing 10.6 ml of H 2 O (18.2 MΩ, Millipore Milli-Q), 6 ml of ammonium hydroxide (28%, VWR International S.A.S.), and 250 ml of ethanol ( 99.9%, Merck) in a 500 ml three-neck flask at room temperature. Next, 33 ml of tetraethyl orthosilicate (98%, Aldrich) was added in the flask quickly, and the mixture was stirred at 700 rpm for 24 h at room temperature. Then the product was centrifuged at 7000 rpm for 1 h, and the translucent solid was collected and sintered at 500 C for 30 min (ramping time 150 min). Preparation of SnO 2 mesoporous single crystals. 1.4 g of SnCl 4.5H 2 O was added to 28 ml of ethanol/water (v/v = 1/1) mixed solvent. After the mixture was stirred for 10 min, 1.4 ml of HCl (37 wt%) was added and stirred for another 10 min. Then 0.8 g of SiO 2 was added in the autoclave and the sealed vessel was heated at 200 C for 12 h in an oven. Afterward, the vessel was cooled to room temperature naturally, and the solid particles settled at the bottom were collected. Then the solid particles were washed with a large amount of water by vacuum filtration. Finally, the silica template was removed by etching in 2 M NaOH at 80 C for 1 h. Then the remaining SnO 2 products were collected by centrifugation and washed with H 2 O and ethanol several times. The SnO 2 MSC paste were deposited from suspension in the ethanol and terpineol (v/v 1:2) with the concentration 5 wt%. Solar cell fabrication. The material synthesis and solar cell fabrication were according to a reported procedure by Michael Gratzel et al. 1 Before depositing CH 3 NH 3 PbI 3, the SnO 2 MSC paste was spin-coated onto patterned fluorine doped tin oxdie (F:SnO 2 ) coated glass (FTO). Then, the film was annealing at 300 C for 30 min. For the TiO 2 coated SnO 2 MSC film, the SnO 2 films were immersed in 50 mm of TiCl 4 aqueous solution at 70 C for 0.5 h and heat-treated at 500 C for 30 min. The TiO 2 nanocyrstals (with 50 nm size diameters) paste were also spin-coated onto the FTO by 3000 rpm and annealing at 500 C for 1 h, the films were also treated in TiCl 4 solution by the method above.

7 PbI 2 (450 mg/ml) in N,N-dimethylformamide (DMF) was spin-coated onto the above films, and then was dipped in a solution of CH 3 NH 3 I in 2-propanol (10 mg/ml) for 20 s and rinsed with 2-propanol. The hole transport layer was used by spin-coating a solution of spiro-meotad, which contains chlorobenzene (130 mg/ml), 26 ul of tert-butylpyridin (TBP) solution and 35 ul of Li-bis(trifluoromethanesulfonyl) imide (Li-TFSI)/acetonitrile (170 mg/ml). For the metal electrode, 50 nm thickness of gold was deposited on the top of the HTM by a thermal evaporation through a metal shadow mask to define the active area of the devices (~7 mm 2 ) and to form a top anode. The cell was packaged by scribbling UV-glue on the top and covered a glass slides, then the cell was exposed to UV light radiation for 10 min to make the glue solidify. The device testing was carried out the glove box after packaging and tested within a metal mask of an aperture (7 mm 2 area). Characterization. Morphologies of the nanomaterials and subsequent nanostructures were directly examined on JEOL6700F SEM at an accelerating voltage of 5 kv. More detailed structural examinations were carried out by transmission electron microscopy (TEM, JEOL 2010F) and high resolution TEM with an accelerating voltage of 200 kv. Powder X-ray diffraction (XRD) patterns were recorded on a Philips high-resolution X-ray diffraction system (model PW1830) with Cu Kα radiation (λ= Å). The light source (Oriel solar simulator, 450 W Xe lamp, AM 1.5 global filter) was calibrated to 1 sun (100 mw cm 2 ) using an optical power meter (Newport, model 1916-C) equipped with a Newport 818P thermopile detector. J V characteristic curves and intensity modulated photocurrent/photovoltage spectroscopy (IMPS/IMVS) were measured by the Zahner controlled intensity modulated photoresponse spectroscopy (C-IMPS) system. We measured the EIS spectra at open voltage and a frequency range from 1Hz and 1MHz with AC amplitude of 10 ma under illumination of simulated solar AM1.5 global light at 100mWcm -2. Z-View Analyst software was used to model the Nyquist plots obtained from the impedance measurements. Incident photon to current conversion efficiencies (IPCEs) was measured on photo current spectra system of CIMPS (CIMPS-PCS) with tunable light source (TLS03).

8

9 Figure S1. Top-sectional SEM images of pristine SnO2 MSCs (A,B) and TiO2-SnO2 MSCs (C,D). Figure S2. Element mapping analysis of (A) TEM image, (B) results of the EDX element analysis, (C) tin and (D) oxygen of pristine SnO2 MSCs.

10 Figure S3. Element mapping analysis of (A) TEM image, (B) results of the EDX element analysis, (C) tin, (D) oxygen and (E) titanium of TiO 2 coated SnO 2 MSCs.

11 Figure S4. Histograms and related Gaussian fits of device parameters measured for 30 devices based on TiO 2 -SnO 2 MSCs perovskite solar cells: (A) J SC, (B) V OC, (C) FF, (D) Efficiency.

12 Figure S5. Histograms and related Gaussian fits of device parameters measured for 30 devices based on pristine SnO 2 MSCs perovskite solar cells: (A) J SC, (B) V OC, (C) FF, (D) Efficiency. Figure S6. The integrated photocurrent densities of the IPCE spectra. SnO 2 MSCs (red line) and TiO 2 -SnO 2 MSCs (black line) cells were measured under AM 1.5 G illumination at 100 mw/cm 2.

13 Figure S7. Stabilized photocurrent density (A) and power conversion efficiency (B) obtained while holding the solar cell near the maximum power point voltage at 0.56V for TiO 2 coated SnO 2 MSC and at 0.35 V for SnO 2 MSC, respectively. Figure S8. Typical photocurrent density versus applied voltage (J~V) curves of the TiO 2 nanocyrstals (NCs) based perovskite solar cells. Note: the TiO 2 NCs based solar

14 cells do not include the TiO 2 dense layer before mesoporous TiO 2 NCs coating onto FTO. After 500 o C annealing, the film was treated by TiCl 4 with the same method as SnO 2 MSCs. Table S1. Average performance parameters of the SnO 2 MSCs and TiO 2 -SnO 2 MSCs based perovskite solar cells. Devices V OC (V) J SC (ma/cm 2 ) FF PCE (%) TiO 2 -SnO 2 MSC ± ± ± ± Pristine SnO 2 MSC ± ± ± ± 0.63 Table S2. Summary of the parameters of Impedance spectroscopy (Figure 4) with fitting the Nyquist plots by the equivalent circuit (inset in Figure 4). Devices R s (Ω) Transport resistance (R 1 )(Ω) Recombination resistance (R 2 )(Ω) TiO 2 -SnO 2 MSCs SnO 2 MSCs TiO 2 nanocrystals Reference (1) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Graetzel, M. Nature, 2013, 499, 316.