Improvement of CH3NH3PbI3 Formation for ... - ACS Publications

Research Article www.acsami.org

Improvement of CH3NH3PbI3 Formation for Efficient and Better Reproducible Mesoscopic Perovskite Solar Cells Changyun Jiang, Siew Lay Lim, Wei Peng Goh, Feng Xia Wei, and Jie Zhang* Institute of Materials Research and Engineering, A*STAR, 3 Research Link, Singapore 117602 S Supporting Information *

ABSTRACT: High-performance perovskite solar cells (PSCs) are obtained through optimization of the formation of CH3NH3PbI3 nanocrystals on mesoporous TiO2 film, using a two-step sequential deposition process by first spin-coating a PbI2 film and then submerging it into CH3NH3I solution for perovskite conversion (PbI2 + CH3NH3I → CH3NH3PbI3). It is found that the PbI2 morphology from different film formation process (thermal drying, solvent extraction, and as-deposited) has a profound effect on the CH3NH3PbI3 active layer formation and its nanocrystalline composition. The residual PbI2 in the active layer contributes to substantial photocurrent losses, thus resulting in low and inconsistent PSC performances. The PbI2 film dried by solvent extraction shows enhanced CH3NH3PbI3 conversion as the loosely packed disk-like PbI2 crystals allow better CH3NH3I penetration and reaction in comparison to the multicrystal aggregates that are commonly obtained in the thermally dried PbI2 film. The as-deposited PbI2 wet film, without any further drying, exhibits complete conversion to CH3NH3PbI3 in MAI solution. The resulting PSCs reveal high power conversion efficiency of 15.60% with a batch-to-batch consistency of 14.60 ± 0.55%, whereas a lower efficiency of 13.80% with a poorer consistency of 11.20 ± 3.10% are obtained from the PSCs using thermally dried PbI2 films. KEYWORDS: perovskite solar cells, mesoscopic structure, perovskite conversion, two-step sequential deposition, reproducibility

■

INTRODUCTION The search for effective usage of the abundant clean energy on earth has encouraged extensive research in solar photovoltaics. Recently, the organic−inorganic hybrid perovskite, methylammonium lead halide MAPbX3 (MA = CH3NH3; X = Cl, Br, I), has been studied extensively as photoharvesting materials by the clean energy research community.1−5 MAPbX3 exhibits long electron/hole diffusion length, high absorption coefficient, and excellent defect tolerance.6−9 A high power conversion efficiency (PCE) perovskite solar cell (PSC) of 20.1% has been certified recently,10,11 as compared to 3.8% when MAPbX3 was first studied as a photoharvester in 2009.12 The high performance, low material cost, and solution processability make PSC an excellent candidate for industry adoption to eventually develop into a central role in solar energy harvesting. A variety of solution-based fabrication processes, such as spincoating, drop-casting, dipping, solvent/vapor extraction, and so on, can be used to fabricate the PSCs with a planar or mesoscopic structure.13−18 CH3NH3PbI3 (MAPbI3) perovskite films are usually formed by either a one-step or a two-step sequential deposition method. In the one-step method, the PbI2 (or PbCl2) and CH3NH3I (MAI) are premixed in a solution before the deposition. The perovskite crystals are formed as the carrier solvent evaporates during the deposition.19−21 In the two-step sequential method, PbI2 is first spin-coated, followed by dipping into MAI solution for the perovskite conversion (PbI2 + CH3NH3I → CH3NH3PbI3).22,23 The two-step sequential method is an ideal approach to mesoscopic PSCs © 2015 American Chemical Society

as it allows control of the perovskite crystallization in the mesoscopic pores and on the TiO2 film surface. However, the two-step process usually leads to incomplete perovskite conversion due to incomplete intercalation of MAI into the PbI2 film.24−26 Therefore, it is more challenging to realize efficient perovskite conversion. Consequently, lower and inconsistent PSC performances are obtained. The recent study by Wu et al. suggested that retarding the crystallization of initially deposited PbI2 film by using dimethyl sulfoxide (DMSO) as solvent could yield more complete perovskite conversion.27 In this study, our focus is to address the PSC fabrication consistency and performance by optimizing the perovskite film quality through controlling the PbI2 layer formation by different postdeposition processes. The selected postdeposition process conditions are thermal drying, solvent extraction by dichloromethane (DCM), and using the as-deposited wet PbI2 film directly. Improved perovskite conversion is obtained by DCMextraction compared to thermal drying. Complete perovskite conversion can be achieved by using as-deposited wet PbI2 film to realize PSCs with higher power conversion efficiency and better reproducibility. Received: August 12, 2015 Accepted: October 22, 2015 Published: October 22, 2015 24726

DOI: 10.1021/acsami.5b07446 ACS Appl. Mater. Interfaces 2015, 7, 24726−24732

Research Article

ACS Applied Materials & Interfaces

Figure 1. Cross-sectional SEM image of a device with layers of FTO, BL-TiO2, mp-TiO2, MAPbI3, HTM (spiro-OMeTAD) and Ag (scale bar: 200 nm).

■

■

EXPERIMENTAL SECTION

RESULTS AND DISCUSSION The mp-TiO2-based PSC was selected as the device structure in this study to take advantage of the large interfacial area between TiO2 and perovskite for efficient charge carrier separation.29,30 The optimized two-step sequential perovskite formation process yielded defect-free solar cells. A cross-sectional SEM image of a typical solar cell device is given in Figure 1. Through controlling the mp-TiO2 thickness and porosity, and the PbI2 film formation process, perovskite crystals filled the pores in the mp-TiO2 film and subsequently formed a capping layer on the surface of the mp-TiO2 layer. The overall photoactive layer (mp-TiO2 + MAPbI3) was ∼530 nm, consisting of ∼250 nm of mp-TiO2:MAPbI3 mesoscopic layer and ∼280 nm of MAPbI3 planar capping layer. All pores in the mp-TiO2 film appeared to be filled by MAPbI3, which is important in realizing highperformance mesoscopic perovskite solar cells.31 The solar cell with the combination of the mesoscopic and planar MAPbI3 layer has unique advantages. The mesoscopic layer provides efficient charge separation at the TiO2/MAPbI3 interface. The complete infiltration of perovskite in the mp-TiO2 layer can also prevent interfacial recombination by ensuring that spiroOMeTAD (hole transporting) is isolated from TiO2 (electron transporting).32 The capping-perovskite layer with larger crystals facilitates efficient charge transport6,7 and provides higher perovskite loading compared to the mp-TiO2:MAPbI3 layer for efficient charge collection and maximum light absorption. Note that the BL-TiO2 between the FTO and the mp-TiO2 layer is not clearly visible from the SEM image. This is due to the thin thickness of 25 nm. The thin BL-TiO2 layer not only limits the resistance to electron transport, but also effectively blocks hole injection from the perovskite into the FTO cathode. This is a prerequisite for achieving high device performance. Through the two-step MAPbI3 formation process, the morphology of PbI2 film, the intermediate product from the first deposition step has profound influence on the final perovskite film quality and composition, thus affecting the PSC performances. To be able to systematically study this effect, three different post treatments on spin-coated PbI2 precursor films were introduced: (a) thermal drying (at 100 °C for 5 min), (b) DCM-extraction by soaking the as-coated films in DCM for 30 s to extract the residual DMF followed by blow drying the films using N2, and (c) using the as-coated wet PbI2 films directly for dipping (as-deposited). All the PbI2 precursor films were prepared using the same spin coating conditions for consistency in the PbI2 precursor loading and the film thickness on mp-TiO2 layer.

A compact TiO2 film of 25 nm was deposited on fluorine-doped tin oxide (FTO) coated substrates (Solaronix, 15 Ω/square) by spray pyrolysis at 300 °C, followed by sintering at 500 °C for 60 min.28 This compact TiO2 layer is also called blocking layer (BL-TiO2). The subsequent mesoporous TiO2 (mp-TiO2) film was formed by spin coating on the BL-TiO2 layer using a TiO2 nanoparticle paste (Dyesol, 18NR-T) at 5000 rpm followed by sintering at 500 °C for 30 min. The TiO2 paste was diluted in ethanol at a ratio of 1:3.5 before use. The resulting mp-TiO2 film had a thickness of 250 nm. The two-step MAPbI3 formation comprised of 1) spin coating of PbI2 film, and 2) dipping the PbI2 film in MAI solution (1% MAI in isopropyl alcohol (IPA)) for 10 min to convert PbI2 into MAPbI3. In the first step, the PbI2 solution (1 M in N,N-dimethylformamide (DMF), heated at 70 °C) was spin coated on the mp-TiO2 substrate at 4000 rpm for 5 s in a N2 purged glovebox. The as-coated PbI2 films were treated by three different processes before dipping in the MAI solution: a) thermal drying at 100 °C for 5 min (thermal drying), b) drying by solvent extraction through soaking as-coated films in DCM for 30 s before blow drying the films using N2 (DCM-extraction), and c) directly using the as coated wet films (as-deposition). During the second step, the PbI2 coated substrate was dipped into MAI solution for 10 min, followed by drying at 70 °C for 10 min to form the perovskite layer. Spiro-OMeTAD was used as the hole transporting material (HTM). Spiro-OMeTAD solution (80 mg/mL in chlorobenzene) containing 35 mM Li[CF3SO2]2N and 200 mM tert-butylpyridine (tBP) was spincoated at 3000 rpm for 40 s on the perovskite layer. Finally, a 150 nm thick Ag layer as the anode was formed by thermal evaporation through a shadow mask to define the active cell area (0.2 cm2). PbI2 and perovskite films on FTO/mp-TiO2 substrates were characterized using X-ray diffraction (XRD), scanning electron microscopy (SEM) and UV−vis spectroscopy. Separately, PbI2 films were prepared for transmission electron microscopy (TEM) characterization by spin coating the PbI2 solution on holely Cu grids, followed by thermal and DCM-extraction drying processes. Photovoltaic current−voltage (J-V) characteristics of solar cell devices were measured in ambient conditions with a Keithley 2400 source-meter unit under 100 mW/cm2 illumination (AM 1.5G) from a 300 W solar simulator (Newport 91160). The intensity of the light source was determined using a Si reference cell (calibrated by Fraunhofer Institute for Solar Energy, ISE) and corrected for spectral mismatch. The fabricated devices were stored in a drybox for an hour before the measurements, without additional conditioning. J−V measurements were performed in forward voltage scan mode (voltage swept from negative to positive), at a moderate scan rate of 50 mV/s. Reverse voltage scans (voltage swept from positive to negative) were also performed on some devices in order to study the hysteresis effects (see the Supporting Information). 24727


Research Article

ACS Applied Materials & Interfaces The perovskite film morphology and crystal structures were studied to investigate the perovskite conversion on PbI2 films processed under different drying conditions. The results are shown in Figure 2. The perovskite crystals have similar cuboid

extraction were evaluated using SEM and TEM. The asdeposited PbI2 film could not be imaged due to its wet and transient nature. As presented in Figure 3, large voids and

Figure 3. SEM images of the PbI2 films formed by (a) thermal drying and (b) DCM-extraction processes (scale bar: 200 nm).

coarse PbI2 domains are observed in the thermally dried PbI2 film (Figure 3a), whereas the DCM-extraction processed PbI2 film has smaller and flakelike domains (Figure 3b). The crosssection images of the PbI2 films can be seen in Figure S1. For TEM analysis, the PbI2 solution was deposited on TEM grids (holely carbon film on Cu grid) and post processed according to thermal drying and DCM-extraction process conditions. Figure 4 shows the dark-field images of the PbI2

Figure 2. (a−c) SEM images and (d) XRD spectra of perovskite films obtained by dipping PbI2 films prepared by thermal drying, DCMextraction and as-deposition processes in MAI solution for 10 min. The XRD spectrum of a PbI2 film is also shown for comparison. The main peaks in the XRD spectra are from PbI2 (#), MAPbI3 (*), and FTO/TiO2 substrate (Δ). PbI2 fraction (wt %) in the PbI2+MAPbI3 composition was determined by Rietveld refinement (see the Supporting Information for details) (SEM scale bar: 100 nm).

Figure 4. TEM images of the PbI2 films prepared on holely TEM Cu grids by (a) thermal drying and (b) DCM-extraction processes (scale bar: 100 nm). Illustrations a1 and b1 depict the crystal packing morphologies of the films.

shapes for all three perovskite films obtained from PbI2 films with different post processing conditions (Figure 2a−c). The sizes of the crystals range from 100 to 250 nm, as determined from SEM images. Slightly larger crystal sizes are observed for the perovskite film obtained from the as-deposited wet PbI2 film (Figure 2c). From the XRD spectra (Figure 2d), a strong crystalline PbI2 peak at 12.6° is observed in the perovskite film processed from the thermally dried PbI2. This indicates a considerable amount of residual PbI2 remaining in the MAPbI3 film, even after dipping in MAI solution for 10 min. The intensity of the residual PbI2 peak decreases in the MAPbI3 film when the DCM-extraction is used to prepare the PbI2 film whereas the peak (at 12.6°) is almost completely eliminated when the as-deposited wet PbI2 film is used for MAPbI3 conversion. After dipping in MAI for 10 min, the weight percentage of PbI2 in the PbI2+MAPbI3 composite in the perovskite films obtained from thermal drying, DCM-extraction and as-deposition processes is 19.0, 4.0, and 175 μm in Solution-Grown CH3NH3PbI3 Single Crystals. Science 2015, 347, 967−970. (9) Nie, W.; Tsai, H.; Asadpour, R.; Blancon, J. C.; Newkirch, A. J.; Gupta, G.; Crochet, J. J.; Chhowalla, M.; Tretiak, S.; Alam, M. A.; Wang, H. L.; Mohite, A. D. High-Efficiency Solution-Processed Perovskite Solar Cells with Millimeter-Scale Grains. Science 2015, 347, 522−525. (10) National Renewable Energy Labs (NREL) Efficiency Chart (2014) < http://www.nrel.gov/ncpv/images/efficiency_chart.jpg>, Accessed 28th May 2015. (11) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. High-Performance Photovoltaic Perovskite Layers Fabricated Through Intramolecular Exchange. Science 2015, 348, 1234−1237. (12) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, 6050−6051. 24731


Research Article

ACS Applied Materials & Interfaces tion Kinetics in Perovskite Solar Cells. ACS Photonics 2015, 2, 589− 594. (31) Leijtens, T.; Lauber, B.; Eperon, G. E.; Stranks, S. D.; Snaith, H. J. The Importance of Perovskite Pore Filling in Organometal Mixed Halide Sensitized TiO2-Based Solar Cells. J. Phys. Chem. Lett. 2014, 5, 1096−1102. (32) Shen, Q.; Ogomi, Y.; Chang, J.; Tsukamoto, S.; Kukihara, K.; Oshima, T.; Osada, N.; Yoshino, K.; Katayama, K.; Toyoda, T.; Hayase, S. Charge Transfer and Recombination at the Metal Oxide/ CH3NH3PbClI2/Spiro-OMeTAD Interfaces: Uncovering the Detailed Mechanism behind High Efficiency Solar Cells. Phys. Chem. Chem. Phys. 2014, 16, 19984−19992. (33) Yan, K.; Long, M.; Zhang, T.; Wei, Z.; Chen, H.; Yang, S.; Xu, J. Hybrid Halide Perovskite Solar Cell Precursors: Colloidal Chemistry and Coordination Engineering behind Device Processing for High Efficiency. J. Am. Chem. Soc. 2015, 137, 4460−4468. (34) Yang, S.; Zheng, Y. C.; Hou, Y.; Chen, X.; Chen, Y.; Wang, Y.; Zhao, H. J.; Yang, H. G. Formation Mechanism of Free standing CH3NH3PbI3 Functional Crystals: In Situ Transformation vs Dissolution−Crystallization. Chem. Mater. 2014, 26, 6705−6710. (35) Liang, K.; Mitzi, D. B.; Prikas, M. T. Synthesis and Characterization of Organic−Inorganic Perovskite Thin Films Prepared Using a Versatile Two-Step Dipping Technique. Chem. Mater. 1998, 10, 403−411. (36) Cottrell, T. L. The Strengths of Chemical Bonds, 2d ed.; Butterworth,: London, 1958.

24732
