Yixin Zhao and Kai Zhu* 1. INTRODUCTION 2. EXPERIMENTAL SECTION

Transcription

1 pubs.acs.org/jpcc CH 3 NH 3 Cl-Assisted One-Step Solution Growth of CH 3 NH 3 PbI 3 : Structure, Charge-Carrier Dynamics, and Photovoltaic Properties of Perovskite Solar Cells Yixin Zhao and Kai Zhu* Chemical and Materials Science Center, National Renewable Energy Laboratory, Denver West Parkway, Golden, Colorado 80401, United States *S Supporting Information ABSTRACT: We demonstrate a novel one-step solution approach to prepare perovskite CH 3 NH 3 PbI 3 films by adding CH 3 NH 3 Cl (or MACl) to the standard CH 3 NH 3 PbI 3 precursor (equimolar mixture of CH 3 NH 3 I and PbI 2 ) solution. The use of MACl strongly affects the crystallization process of forming pure CH 3 NH 3 PbI 3, leading not only to enhanced absorption of CH 3 NH 3 PbI 3 but also to significantly improved coverage of CH 3 NH 3 PbI 3 on a planar substrate. Compared to the standard one-step solution approach for CH 3 NH 3 PbI 3, using MACl improves the performance of CH 3 NH 3 PbI 3 solar cells from about 2% to 12% for the planar cell structure and from about 8% to 10% for the mesostructured device architecture. Although we find no significant effect of using MACl on charge transport and recombination in mesostructured perovskite cells, the recombination resistance for planar cells increases by 1 2 orders of magnitude by using MACl. These results suggest that this new one-step solution approach is promising for controlling CH 3 NH 3 PbI 3 growth to achieve high-performance perovskite solar cells. 1. INTRODUCTION Methylammonium lead halide perovskites represent a novel class of absorbers for solar conversion applications. 1,2 Solar conversion efficiencies over 15% have recently been demonstrated by multiple groups. 3,4 Various perovskite absorbers (e.g., CH 3 NH 3 PbI 3, CH 3 NH 3 PbI 3 x Cl x, and CH 3 NH 3 PbBr 3 ) and device architectures (e.g., mesoporous and planar cell configurations) have been examined with promising results by using solution processing or thermal evaporation In general, CH 3 NH 3 PbI 3 has demonstrated great success for mesostructured cells using solution processing, 4,16 whereas CH 3 NH 3 PbI 3 x Cl x has shown promise for constructing efficient planar devices by evaporation. 3,11 These materials have also inspired fundamental studies on transport and recombination properties by several groups Controlling the morphology and crystal structure of perovskite absorbers is important for achieving high-performance devices. 22,23 For solution processing, this is often done by adjusting the concentration and solvent of precursors, spin speed/duration, and postgrowth annealing temperature/duration/atmosphere. In contrast to the success of CH 3 NH 3 PbI 3 x Cl x for planar devices, solution processing of CH 3 NH 3 PbI 3 has mainly been used in mesoporous cell structures. 4,16,24 The challenge is to prepare uniform, pinhole-free CH 3 NH 3 PbI 3 films on planar substrate by conventional solution deposition for making efficient planar devices. 25 The standard CH 3 NH 3 PbI 3 precursor uses the mixture of CH 3 NH 3 I and PbI 2, 24 whereas the standard CH 3 NH 3 PbI 3 x Cl x precursor consists of CH 3 NH 3 I and PbCl 2 with a molar ratio of 3:1. 7 A recent study addresses the challenge of preparing CH 3 NH 3 PbI 3 on a planar substrate by using a vapor-assisted two-step solution approach, 14 which, however, requires a relatively long (2 4 h) exposure of PbI 2 to the CH 3 NH 3 I vapor during the second step. Herein, we present a new one-step solution approach by introducing CH 3 NH 3 Cl (or MACl) to the standard CH 3 NH 3 PbI 3 precursor (equimolar mixture of CH 3 NH 3 I and PbI 2 ) solution to prepare perovskite CH 3 NH 3 PbI 3 on both mesoporous and planar TiO 2 substrates. Optical and structural characterizations show that the use of MACl adjusts the crystallization process for forming CH 3 NH 3 PbI 3. Depending on the amount of MACl in the precursor solution, the optimum crystallization process takes place from a few minutes to several tens of minutes. We find that using MACl not only improves absorption of CH 3 NH 3 PbI 3 but also enhances the CH 3 NH 3 PbI 3 coverage on planar substrates, leading to significant improvement of device performance. Charge transport and recombination properties are examined by intensity-modulated photocurrent/photovoltage spectroscopies (IMPS/IMVS) and impedance spectroscopy (IS). 2. EXPERIMENTAL SECTION Transparent Conducting Substrate and Mesoporous TiO 2 Film. Fluorine-doped transparent conducting SnO 2 - coated glass substrate (FTO; TEC15, Hartford, IN) was prepatterned by etching with Zn powder and 25 wt % HCl Received: March 18, 2014 Revised: April 16, 2014 Published: April 16, American Chemical Society 9412 dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

2 The Journal of Physical Chemistry C solution for about 2 min. The patterned FTO substrate was then cleaned by soaking in 5 wt % NaOH in alcohol for 16 h and then rinsing it sequentially with deionized (DI) water and ethanol. The cleaned FTO substrate was subsequently coated with a compact TiO 2 layer by spray pyrolysis using 0.2 M Ti(IV) bis(ethyl acetoacetate) diisopropoxide in 1-butanol solution at 450 C, followed by annealing at 450 C for 1 h. 26 The 20 nm TiO 2 nanoparticles were synthesized by following previous reports. 27 The TiO 2 paste with 6 wt % of the 20 nm TiO 2 nanoparticles mixed with terpineol and ethylcellulose were screen-printed with 3 μm emulsion thickness on the patterned FTO substrates. The printed mesoporous TiO 2 film was annealed at 500 C for 0.5 h. The average film thickness was determined by an Alpha-Step 500 surface profiler. The TiO 2 films were then soaked in 0.04 M TiCl 4 solution at 65 C for 30 min, followed by rinsing with DI water and ethanol, and finally dried under N 2. The TiCl 4 -treated TiO 2 films (for mesostructured cells) and compact TiO 2 films (for planar devices) on patterned FTO were annealed again at 500 C for 30 min before the deposition of perovskite CH 3 NH 3 PbI 3. Device Fabrication. CH 3 NH 3 I (MAI) was synthesized by reacting methylamine (33 wt % ethanol solution) and hydroiodic acid (57 wt % in water, Aldrich) with the molar ratio of 1.2:1 in an ice bath for 2 h with stirring followed by vacuum drying and cleaning with ethyl acetate. CH 3 NH 3 Cl (MACl) was synthesized by reacting methylamine (33 wt % ethanol solution) and 33 wt % hydrocholoride acid with the molar ratio of 1.2:1 in an ice bath for 2 h with stirring followed by vacuum drying and cleaning with acetonitrile g of PbI 2 (1.5 mmol), g of MAI (1.5 mmol), and 0 g, g (0.75 mmol), g (1.5 mmol), or g (3 mmol) of MACl was dissolved in 2.75 g of dimethylformamide (DMF) at room temperature to form four different CH 3 NH 3 PbI 3 precursor solutions noted as 0 MACl, 0.5 MACl, 1 MACl, and 2 MACl, respectively. Devices were fabricated in ambient condition (unless stated otherwise) as detailed below. The perovskite CH 3 NH 3 PbI 3 precursor solutions were spin-coated onto (1) 650 nm thick TiO 2 mesoporous films on FTO at 3000 rpm for 30 s for mesostructured cells and (2) compact TiO 2 films on FTO at 2500 rpm for 10 s. The perovskite-coated films were then annealed on a hot plate at 100 C for about 5 45 min. For both mesostructured and planar perovskite solar cells, a hole-transport material (HTM) solution was spin-coated on the perovskite-covered TiO 2 electrodes at 4000 rpm for 30 s. The HTM solution consists of 0.1 M 2,2,7,7 -tetrakis(n,n-dip-methoxyphenylamine)-9,9 -spirobifluorene (spiro-meo- TAD), M bis(trifluoromethane)sulfonimide lithium salt (Li-TFSi), and 0.12 M 4-tert-butylpyridine (tbp) in chlorobenzene/acetonitrile (10:1, v/v) solution. Finally, a 150 nm thick Ag layer was deposited on the HTM layer by thermal evaporation. The active area of each device was about cm 2. Characterization. The crystal structures of the perovskite films were measured by X-ray diffraction (XRD, Rigaku D/Max 2200 diffractometer with Cu Kα radiation). The absorption spectra of the mesoporous and planar perovskite films were characterized by an UV/vis NIR spectrophotometer (Cary- 6000i). The photocurrent voltage (J V) characteristic of perovskite CH 3 NH 3 PbI 3 solar cells were measured with a Keithley 2400 source meter under the simulated AM 1.5G illumination (100 mw/cm 2 ; Oriel Sol3A Class AAA Solar Simulator), from open circuit to short circuit with a scan rate of 0.2 V/s. J V hysteresis was observed when scanned from short circuit to open circuit, especially for planar devices. The reason for the stronger hysteresis for planar perovskite cells is under further investigation. Charge transport and recombination properties of the mesostructured perovskite cells were measured by intensity-modulated photocurrent and photovoltage spectroscopies as described previously. 28 Impedance spectroscopy (IS) was done using a PARSTAT 2273 workstation with the frequency range of 0.1 Hz 100 khz and the modulation amplitude of 10 mv. The IS spectra were analyzed using ZView 2.9c software (Scribner Associates). 3. RESULTS AND DISCUSSION Figure 1 shows the effect of varying annealing time (1 45 min) at 100 C on the appearance of perovskite CH 3 NH 3 PbI 3 films Figure 1. Images of the perovskite films (on mesoporous TiO 2 ) prepared from CH 3 NH 3 PbI 3 precursors with different amounts of CH 3 NH 3 Cl (or MACl; as indicated) and annealed at 100 C with varying duration (as indicated). See text for details on the precursor compositions. prepared by using the perovskite precursors with different amounts of added CH 3 NH 3 Cl (MACl). The perovskite precursor solution contains PbI 2, CH 3 NH 3 I (or MAI), and MACl with a molar ratio of 1:1:x (x varies from 0 to 2; the maximum dissoluble x value is about 2.8). The thickness of the mesoporous TiO 2 film is about 650 nm. When no MACl is used, the perovskite film turns brown within 1 min of annealing at 100 C. The absorbance of this film stays virtually unchanged at 5 min and then decreases when annealed for 10 min (Figure S1a in the Supporting Information). With an increasing amount of MACl, the process that turns the perovskite film brown/dark brown with annealing occurs at significantly slower rates. When 2 MACl is used, it takes more than 25 min for the perovskite film to turn brown. The annealing-time-dependent absorption spectra for these perovskite films prepared using different amounts of MACl are shown in Figure S1 (Supporting Information). It is worth noting that adding MACl to the precursor solution not only slows down the perovskite film darkening process but also darkens the color the final perovskite films (the color of the final perovskite film changes from brown with no MACl to dark brown with 1 2 MACl; Figure 1). Figure 2a compares the X-ray diffraction (XRD) patterns of CH 3 NH 3 PbI 3 on mesoporous TiO 2 film prepared from precursor solutions containing different amounts of MACl. When the absorption spectrum reaches about maximum for each amount of added MACl (about 5 min for 0 MACl, 10 min for 0.5 MACl, 25 min for 1 MACl, and 45 min for 2 MACl), the perovskite samples show the same XRD patterns for CH 3 NH 3 PbI 3 as we reported previously. 29,30 It is interesting to note that the strongest CH 3 NH 3 PbI 3 (110) peak near dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

3 The Journal of Physical Chemistry C Figure 2. (a) XRD patterns of CH 3 NH 3 PbI 3 on mesoporous TiO 2 film as a function of the MACl amount in the precursors for different annealing time at 100 C. Magnified views of the XRD patterns (b) near 14 for perovskite films using 0 2 MACl with their respective optimum annealing time and (c) for the 2-MACl sample with different annealing time (1 45 min). Peaks associated with the perovskite CH 3 NH 3 PbI 3 structure are labeled. The other peaks are either from the TiO 2 /FTO substrate or from intermediate structure related to MACl. increases slightly in intensity and also becomes narrower with the amount of MACl added to the precursor solution (Figure 2b). This may suggest an increase of the crystallinity or a change of the preferred orientation of perovskites with higher MACl concentration, which is presumably caused by the slower crystallization process at higher MACl concentration. The perovskite film using 2 MACl displays the longest time for complete color change, and its XRD patterns as a function of the annealing time (1 45 min) at 100 C are also shown in Figure 2a. The main characteristic CH 3 NH 3 PbI 3 (110) peak near 14 only appears after 25 min annealing. At the early stage of annealing (1 10 min), we observe several intermediate peaks (Figure 2c), which all disappear after annealing for 45 min. Table 1 shows the energy dispersive X-ray (EDX) analysis Table 1. Effect of MACl Amount (x) and Annealing Time on the Pb:I:Cl Ratio of Perovskite Films x MACl (time) Pb I Cl 2 (1 min) (0.3 a ) 1.6 (0.2) 2 (10 min) (0.3) 0.6 (0.1) 2 (25 min) (0.3) 0.1 (0.1) 2 (45 min) (0.3) 1 (25 min) (0.3) 0.5 (10 min) (0.4) 0 (5 min) (0.4) a The errors of the element ratios are obtained based on the EDX detection limit of 1%. of the perovskite films prepared from the 2 MACl precursor solution, suggesting that a significant amount of Cl is initially incorporated into the film with an I:Cl ratio of 2.7:1.6. The amount of Cl detected in the perovskite film decreases rapidly from 1.6 to 0.1 with annealing time changing from 1 to 25 min. Longer annealing time (45 min) leads to a complete loss of Cl (within the EDX detection limit of 1%). The intermediate XRD peaks for the 2 MACl samples with short annealing times (1 10 min) are presumably associated with the incorporation of MACl or its variation, which presumably sublimes from the film with longer annealing time. This observation is similar to a recent study on the formation of mixed halide CH 3 NH 3 PbI 3 x Cl x from the precursor containing CH 3 NH 3 I and PbCl 2 with a molar ratio of 3:1. 31 The significant structural change for the 2 MACl sample based on XRD measurement is consistent with the changes of its absorption spectra as a function of annealing time (Figure S1d). Similar to the 2-MACl sample, no Cl is observed for other samples using zero or the lower amounts of MACl with proper annealing times (Table 1). These results suggest that adding MACl to the CH 3 NH 3 PbI 3 precursor solution could form some intermediate crystal structure related to MACl, which slows down the crystallization of CH 3 NH 3 PbI 3. The intermediate structure disappears with prolonged annealing to finally form perovskite CH 3 NH 3 PbI 3. Figure 3 shows the typical scanning electron microscopy (SEM) images of top views of the annealed perovskite films prepared from precursor solutions with different amounts of MACl. Images a d correspond to the perovskite films deposited on mesoporous TiO 2 films (650 nm thickness) using 0 2 MACl, respectively. When no MACl is used, some islands of submicrometer-sized CH 3 NH 3 PbI 3 are observed on the top surface of the mesoporous TiO 2 film. 32 These islands either disappear or become less obvious when MACl is used. Despite the slight difference of the perovskite appearance on the top surface of TiO 2 films, the overall appearances of these four films are similar and basically adopt the structure of the underlying mesoporous TiO 2 films. In contrast, adding MACl to the precursor solution leads to dramatic changes of the morphology of the perovskite CH 3 NH 3 PbI 3 films deposited directly on compact TiO 2 on the fluorine-doped tin oxide (FTO) glass substrate. Images e h correspond to the perovskite films deposited on compact TiO 2 using 0 2 MACl, respectively. When no MACl is used, large elongated crystal plates are formed, with a significant portion of the substrate being exposed without CH 3 NH 3 PbI 3 coverage. When 0.5 MACl is used, the large crystal plates disappear partially and some small crystals start to form, leading to enhanced surface coverage of perovskites. When using more MACl (x = 1 and 2), the elongated large crystal plates totally disappear and the substrate is coated with interconnected relatively small crystals with a high surface coverage. The difference in the surface coverage of perovskite films on the planar substrate will likely affect the device characteristics. 22 Figure 4a shows the effect of using MACl on the current density voltage (J V) characteristics of mesostructured perovskite solar cells based on 650 nm thick TiO 2 mesoporous films. The cell with 0 MACl demonstrates a short-circuit photocurrent density (J sc ) of ma/cm 2, open-circuit voltage (V oc ) of V, and fill factor (FF) of to yield an efficiency (η) of 8.23%. The cell efficiency increases to about 9% 10% when MACl (x = 0.5 2) is used to adjust the crystallization process of CH 3 NH 3 PbI 3. The efficiency improvement results mainly from the higher J sc associated with stronger light absorption (Figure S1) by using MACl. When CH 3 NH 3 PbI 3 is deposited directly on the planar compact TiO 2, the device performance (Figure 4b) shows a similar trend with MACl compared to the mesostructured perovskite solar cells. However, the degree of performance improvement by MACl for planar devices is much more significant than that for 9414 dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

4 The Journal of Physical Chemistry C Figure 3. Typical SEM images of perovskite CH3NH3PbI3 grown on (a d) mesoporous TiO2 film and (e h) planar TiO2 compact layer with different MACl amount (a, e: 0; b, f: 0.5; c, g: 1; d, h: 2). 12% (1 2 MACl). The device parameters (Jsc, Voc, FF, and η) for all these mesoporous and planar cells are given in Table S1 (Supporting Information). Impedance spectroscopy13,19,24,33 is used to study the effect of using MACl on the recombination resistance (Rrec) for mesostructured and planar perovskite CH3NH3PbI3 solar cells. Figrue 5a shows that the Rrec values as a function of voltage for mesostructured perovskite solar cells. The Rrec for all mesostructured perovskite cells depends strongly with the bias voltage, following an approximately exponential decrease with voltage. This voltage dependence of Rrec for mesostructured cells is in agreement with previous reports.13,19,24 It is noteworthy that all mesostructured perovskite cells exhibit essentially the same voltage dependence of Rrec, suggesting that using MACl in the precursor has no significant effect on the recombination kinetics in mesostructured cells. In contrast, we observe a strong MACl effect on the voltage dependence of Rrec for planar perovskite cells (Figure 5b). The Rrec values for the planar device not using MACl are generally 1 2 orders of magnitude lower than those for the planar devices using MACl to assist the crystallization of CH3NH3PbI3. Thus, the recombination rate for planar samples prepared without MACl is much faster than that for planar samples prepared with MACl. The observed difference in Rrec for these planar devices is consistent with their markedly different dark J V characteristics (Figure S2 in the Supporting Information). The onset voltage of the dark current shifts from about 500 mv to over 700 mv when MACl is used. Both Rrec and dark J V results are consistent with the influence of MACl on the morphologies of planar perovskite CH3NH3PbI3 films grown directly on the compact TiO2 layer (images e h in Figure 3). Figure 4. Effect of MACl on the J V curves of (a) mesostructured and (b) planar perovskite CH3NH3PbI3 solar cells. All devices use spiromeotad as the hole conductor and Ag as the top contact (details in the Experimental Section). the mesostructured devices. The Jsc increases substantially from about 6 to 18 ma/cm2 when the amount of MACl is changed from 0 to 0.5. When 1 2 MACl is used, the Jsc further increases to >20 ma/cm2. The Voc increases from about 0.8 V for 0 MACl to >1 V for MACl. The FF increases from about 0.41 to a range of As a result, the overall efficiency is improved from about 2% (0 MACl) to 11% (0.5 MACl) to Figure 5. Effect of MACl on the recombination resistance (Rrec) as a function of voltage for (a) mesostructured and (b) planar perovskite CH3NH3PbI3 solar cells dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

5 The Journal of Physical Chemistry C When no MACl is used, a significant portion of the substrate is exposed without CH 3 NH 3 PbI 3 coverage, which could lead to enhanced recombination. To help understand the role of MACl on the perovskite film formation using our new precursor compositions (i.e., mixture of PbI 2, MAI, and MACl with different molar ratios), we examined the perovskite films prepared from the precursor containing only PbI 2 and MACl in the absence of MAI. For the purpose of discussion, we name this perovskite film as PbI 2 - MACl. Figure 6 shows the effect of annealing time (at 100 C) and MACl is clear and stable in ambient condition. These results suggest that the exact precursor composition is critical to the formation of perovskite films. Thus, the success in preparing CH 3 NH 3 PbI 3 film in this study cannot be associated with PbI 2 -MACl in the absence of MAI. Here we hypothesize that the CH 3 NH 3 PbI 3 film with MACl-added CH 3 NH 3 PbI 3 precursors could be formed through a possible intermediate, MAI PbI 2 xmac1, with unknown crystal structures. The additive MACl works as a sacrificial agent to form this intermediate and then slowly decomposes and sublimes during the annealing process. We further examined the mixed halide CH 3 NH 3 PbI 3 x Cl x prepared from the precursor containing a mixture of MAI and PbCl 2 with a molar ratio of 3:1. 7 The Pb concentration of this precursor is the same as our 2-MACl precursor solution. Figure 7a shows the SEM image of the CH 3 NH 3 PbI 3 x Cl x film Figure 6. Effect of annealing time (at 100 C) on the XRD patterns of perovskite films prepared from precursor containing equimolar PbI 2 and MACl. In addition to the main perovskite (110) peak near 14, two new peaks appear at about 12.6 (denoted by pound) and 15.5 (denoted by asterisk). on the XRD patterns of the PbI 2 -MACl films prepared from the stoichiometric solution of PbI 2 and MACl with the same Pb concentration as used in our new MACl-added CH 3 NH 3 PbI 3 precursors. After 1 min annealing, the PbI 2 -MACl film displays the characteristic perovskite (110) peak near 14. There are two additional new XRD peaks at about 12.6 and 15.5, respectively. In contrast, the perovskite film using MACl-added CH 3 NH 3 PbI 3 precursor exhibits different XRD patterns during the early stage of annealing (Figure 2). The peak near 12.6 is ascribed to the formation of PbI 2, whereas the peak near 15.5 can be attributed to the formation of CH 3 NH 3 PbCl With increasing annealing time, the intensities of both the 14 and 15.5 perovskite peaks decrease. In contrast, the 12.6 PbI 2 peak increases with annealing time. With annealing time longer than 50 min, the two perovskite peaks near 14 and 15.5 disappear completely. These XRD results suggest that the PbI 2 - MACl film is not stable and decomposes to primarily PbI 2 with annealing. Consistent with the XRD results, the absorption spectrum of the PbI 2 -MACl film also undergoes significant changes with annealing time (Figure S6). The initial absorption spectrum of the PbI 2 -MACl film is similar to that of CH 3 NH 3 PbI 3 with a brown color. The film turns yellowish after >20 min annealing, and its absorption spectrum becomes dominated by the absorption of PbI 2, which is consistent with the XRD results. Devices based on the PbI 2 -MACl films with 5 10 min annealing only show 2 3% efficiencies. These results suggest that direct reaction between PbI 2 and MACl (in the absence of MAI) cannot lead to the formation of pure and good CH 3 NH 3 PbI 3 perovskites. Furthermore, we found that the maximum molar ratio of MACl:PbI 2 in the PbI 2 -MACl precursor solutions is 1:1. Using a larger molar ratio for MACl:PbI 2 cannot lead to a clear, fully dissolved precursor solution. In contrast, our new precursor containing PbI 2, MAI, Figure 7. (a) SEM image, (b) XRD patterns, and (c) EDX analysis of the mixed halide CH 3 NH 3 PbI 3 x Cl x prepared from the precursor containing MAI and PbCl 2 (3:1 molar ratio). annealed at 100 C for 45 min. The overall film morphology is similar to the CH 3 NH 3 PbI 3 films prepared using the CH 3 NH 3 PbI 3 precursors with the addition of 1 2 MACl (Figure 3g,h). The crystal structure evolution of the CH 3 NH 3 PbI 3 x Cl x film was examined by the annealing time dependence of the XRD patterns. With 1 min annealing at 100 C, the film exhibit three XRD peaks near 12.6,14, and 15.5, which are similar to the XRD patterns of the PbI 2 -MACl films (Figure 6) during the early stage of annealing. Again, the perovskite film using MACl-added CH 3 NH 3 PbI 3 precursors exhibits different XRD patterns during the initial annealing stage, implying the existence of different crystal structures formed during the early stage of annealing. With increasing annealing time to 45 min, only the perovskite peak near 14 is left for the CH 3 NH 3 PbI 3 x Cl x film, similar to the XRD patterns for CH 3 NH 3 PbI 3 shown in Figure 2. Consistent with this observation, the EDX analysis (Figure 7c) of the CH 3 NH 3 PbI 3 x Cl x film shows that with increasing annealing time, the Cl:I ratio decreases from about 1.6:2.7 at 1 min to 9416 dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

6 The Journal of Physical Chemistry C 0.5:2.8 at 20 min. The CH 3 NH 3 PbI 3 x Cl x film annealed for 45 min does not show any detectable trace of Cl. Both XRD and EDX results raise a question whether the commonly called mixed halide CH 3 NH 3 PbI 3 x Cl x is actually CH 3 NH 3 PbI 3. A recent paper by Graẗzel and co-workers also raised the same question based on annealing-temperature-dependent XRD studies of the CH 3 NH 3 PbI 3 x Cl x films. 31 However, other reports do show clear evidence of the presence of Cl in the CH 3 NH 3 PbI 3 x Cl x films. 7,18,34 In view of these different results from different groups, one should be more careful in the assignment of CH 3 NH 3 PbI 3 x Cl x or CH 3 NH 3 PbI 3 for specific perovskites in the future. For the purpose of discussion in this study, we will use CH 3 NH 3 PbI 3 x Cl x to describe the perovskites prepared from the standard 3MAI-PbCl 2 precursor. The J V curve for a planar CH 3 NH 3 PbI 3 x Cl x perovskite cell is shown in Figure S7 (Supporting Information). The cell efficiency is 11.86% with a J sc of ma/cm 2, V oc of V, and FF of The performance of the planar CH 3 NH 3 PbI 3 x Cl x perovskite cell is similar to that of the planar CH 3 NH 3 PbI 3 cells prepared from the CH 3 NH 3 PbI 3 precursors with 1 2 MACl additives in this study. Figure 8 Figure 8. Recombination resistance R rec as a function of voltage for planar perovskite solar cell based on the mixed halide CH 3 NH 3 PbI 3 x Cl x prepared from the precursor containing MAI and PbCl 2 (3:1 molar ratio). The R rec values for the planar CH 3 NH 3 PbI 3 cells prepared from the CH 3 NH 3 PbI 3 precursor with 2 MACl additive are also plotted for comparison. shows the recombination resistance R rec as a function of voltage for the planar CH 3 NH 3 PbI 3 x Cl x solar cell. The R rec values for the planar CH 3 NH 3 PbI 3 cells prepared with 2-MACl additive are also plotted for comparison. It is apparent that these two samples follow almost identical voltage dependence of the recombination resistances. Taken together the results of SEM, XRD, J V, and impedance spectroscopy, it is reasonable to believe that the CH 3 NH 3 PbI 3 x Cl x film prepared from the 3MAI-PbCl 2 precursor is similar to the CH 3 NH 3 PbI 3 film prepared from the MACl-added CH 3 NH 3 PbI 3 precursors. Despite these similarities, we also noticed significant differences between these two solution approaches. (1) First, the crystal structures of the perovskite films prepared with these two approaches during the early stage of annealing are different (Figures 2 and 7), indicating that the exact precursor composition is critical to the process of perovskite formation. (2) We also found that the ratio of MAI:PbCl 2 in the precursor for preparing CH 3 NH 3 PbI 3 x Cl x cannot be less than 3; otherwise, PbCl 2 cannot be fully dissolved. This is probably the reason why the CH 3 NH 3 PbI 3 x Cl x precursor has a molar ratio of 3:1 for MAI:PbCl 2. In contrast, our approach adds a wide range of amounts of MACl to the standard CH 3 NH 3 PbI 3 precursor (MAI and PbI 2 ), allowing an enhanced flexibility to control the kinetics of the perovskite crystallization process. (3) Moreover, we found that the 3MAI-PbCl 2 precursor solution in our trials turns from green-yellow to brown-yellow after 1 week storage in air. In contrast, the PbI 2 -MAI-2MACl precursor solution (with the same Pb concentration as in the 3MAI-PbCl 2 precursor solution) stays clear with a green-yellow color for more than three months when stored in air. 4. CONCLUSIONS In summary, we have developed a one-step solution approach to prepare perovskite CH 3 NH 3 PbI 3 on a mesoporous TiO 2 film or on a planar, compact TiO 2 layer on FTO. In this new synthetic approach, CH 3 NH 3 Cl (or MACl) is added to the standard CH 3 NH 3 PbI 3 precursor (equimolar mixture of CH 3 NH 3 I and PbI 2 ) solution to adjust the crystallization process for CH 3 NH 3 PbI 3. Depending on the amount of MACl used in the precursor solution, the optimum crystallization time for forming pure CH 3 NH 3 PbI 3 with the strongest absorption varies from a few minutes to several tens of minutes. The use of MACl not only leads to enhanced absorption of CH 3 NH 3 PbI 3 but also to significantly improved coverage of CH 3 NH 3 PbI 3 on a planar substrate. Compared to the standard one-step solution approach for CH 3 NH 3 PbI 3, using MACl improves the performance of perovskite solar cells. For the mesostructured device architecture, the efficiency is enhanced from about 8% to 10%, whereas for the planar cell structure, the efficiency is improved from about 2% to 12%. Although no significant dependence on MACl is found for charge transport and recombination in mesostructured perovskite cells, the recombination resistance for planar cells increases by 1 2 orders of magnitude by using MACl. The significant performance improvement for planar perovskite cells is attributed primarily to the improved morphology of the perovskite films resulting from the use of MACl to control the crystallization process for forming CH 3 NH 3 PbI 3. ASSOCIATED CONTENT *S Supporting Information Absorption spectra, solar cell parameters, dark J V curves, typical IS spectra, IMPS/IMVS, and SEM images. This material is available free of charge via the Internet at AUTHOR INFORMATION Corresponding Author * Kai.Zhu@nrel.gov. Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS We acknowledge the support by the U.S. Department of Energy/National Renewable Energy Laboratory s Laboratory Directed Research and Development (LDRD) program under Contract DE-AC36-08GO K.Z. acknowledges the support on the charge transport and recombination studies by the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences, U.S. Department of Energy, under Contract DE-AC36-08GO28308 with the National Renewable Energy Laboratory dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

7 The Journal of Physical Chemistry C REFERENCES (1) Park, N. G. Organometal Perovskite Light Absorbers Toward a 20% Efficiency Low-Cost Solid-State Mesoscopic Solar Cell. J. Phys. Chem. Lett. 2013, 4, (2) Snaith, H. J. Perovskites: The Emergence of a New Era for Low- Cost, High-Efficiency Solar Cells. J. Phys. Chem. Lett. 2013, (3) Liu, M.; Johnston, M. B.; Snaith, H. J. Efficient Planar Heterojunction Perovskite Solar Cells by Vapour Deposition. Nature 2013, 501, (4) Burschka, J.; Pellet, N.; Moon, S. J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Gratzel, M. Sequential Deposition as a Route to High-Performance Perovskite-Sensitized Solar Cells. Nature 2013, 499, (5) Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G. 6.5% Efficient Perovskite Quantum-Dot-Sensitized Solar Cell. Nanoscale 2011, 3, (6) Abu Laban, W.; Etgar, L. Depleted Hole Conductor-Free Lead Halide Iodide Heterojunction Solar Cells. Energy Environ. Sci. 2013, 6, (7) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites. Science 2012, 338, (8) Cai, B.; Xing, Y.; Yang, Z.; Zhang, W.-H.; Qiu, J. High Performance Hybrid Solar Cells Sensitized by Organolead Halide Perovskites. Energy Environ. Sci. 2013, 6, (9) Qiu, J.; Qiu, Y.; Yan, K.; Zhong, M.; Mu, C.; Yan, H.; Yang, S. All- Solid-State Hybrid Solar Cells Based on a New Organometal Halide Perovskite Sensitizer and One-Dimensional TiO 2 Nanowire Arrays. Nanoscale 2013, 5, (10) Noh, J. H.; Im, S. H.; Heo, J. H.; Mandal, T. N.; Seok, S. I. Chemical Management for Colorful, Efficient, and Stable Inorganic Organic Hybrid Nanostructured Solar Cells. Nano Lett. 2013, 13, (11) Docampo, P.; Ball, J. M.; Darwich, M.; Eperon, G. E.; Snaith, H. J. Efficient Organometal Trihalide Perovskite Planar-Heterojunction Solar Cells on Flexible Polymer Substrates. Nat. Commun. 2013, 4, (12) Bi, D.; Yang, L.; Boschloo, G.; Hagfeldt, A.; Johansson, E. M. J. Effect of Different Hole Transport Materials on Recombination in CH 3 NH 3 PbI 3 Perovskite-Sensitized Mesoscopic Solar Cells. J. Phys. Chem. Lett. 2013, 4, (13) Christians, J. A.; Fung, R. C. M.; Kamat, P. V. An Inorganic Hole Conductor for Organo-Lead Halide Perovskite Solar Cells. Improved Hole Conductivity with Copper Iodide. J. Am. Chem. Soc. 2014, 136, (14) Chen, Q.; Zhou, H.; Hong, Z.; Luo, S.; Duan, H.-S.; Wang, H.- H.; Liu, Y.; Li, G.; Yang, Y. Planar Heterojunction Perovskite Solar Cells via Vapor-Assisted Solution Process. J. Am. Chem. Soc. 2013, 136, (15) Edri, E.; Kirmayer, S.; Cahen, D.; Hodes, G. High Open-Circuit Voltage Solar Cells Based on Organic Inorganic Lead Bromide Perovskite. J. Phys. Chem. Lett. 2013, 4, (16) Jeon, N. J.; Lee, J.; Noh, J. H.; Nazeeruddin, M. K.; Graẗzel, M.; Seok, S. I. Efficient Inorganic Organic Hybrid Perovskite Solar Cells Based on Pyrene Arylamine Derivatives as Hole-Transporting Materials. J. Am. Chem. Soc. 2013, 135, (17) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Gratzel, M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- and Hole-Transport Lengths in Organic-Inorganic CH 3 NH 3 PbI 3. Science 2013, 342, (18) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J. Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite Absorber. Science 2013, 342, (19) Kim, H. S.; Mora-Sero, I.; Gonzalez-Pedro, V.; Fabregat- Santiago, F.; Juarez-Perez, E. J.; Park, N. G.; Bisquert, J. Mechanism of Carrier Accumulation in Perovskite Thin-Absorber Solar Cells. Nat. Commun. 2013, 4, (20) Juarez-Perez, E. J.; Wuβler, M.; Fabregat-Santiago, F.; Lakus- Wollny, K.; Mankel, E.; Mayer, T.; Jaegermann, W.; Mora-Sero, I. Role of the Selective Contacts in the Performance of Lead Halide Perovskite Solar Cells. J. Phys. Chem. Lett. 2014, 5, (21) Colella, S.; Mosconi, E.; Fedeli, P.; Listorti, A.; Gazza, F.; Orlandi, F.; Ferro, P.; Besagni, T.; Rizzo, A.; Calestani, G.; et al. MAPbI 3 x Cl x Mixed Halide Perovskite for Hybrid Solar Cells: The Role of Chloride as Dopant on the Transport and Structural Properties. Chem. Mater. 2013, 25, (22) Eperon, G. E.; Burlakov, V. M.; Docampo, P.; Goriely, A.; Snaith, H. J. Morphological Control for High Performance, Solution- Processed Planar Heterojunction Perovskite Solar Cells. Adv. Funct. Mater. 2014, 24, (23) Choi, J. J.; Yang, X.; Norman, Z. M.; Billinge, S. J. L.; Owen, J. S. Structure of Methylammonium Lead Iodide Within Mesoporous Titanium Dioxide: Active Material in High-Performance Perovskite Solar Cells. Nano Lett. 2013, 14, (24) Kim, H.-S.; Lee, C.-R.; Im, J.-H.; Lee, K.-B.; Moehl, T.; Marchioro, A.; Moon, S.-J.; Humphry-Baker, R.; Yum, J.-H.; Moser, J. E.; et al. Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (25) Kim, H.-S.; Im, S. H.; Park, N.-G. Organolead Halide Perovskite: New Horizons in Solar Cell Research. J. Phys. Chem. C 2014, 118, (26) Jang, S. R.; Zhu, K.; Ko, M. J.; Kim, K.; Kim, C.; Park, N. G.; Frank, A. J. Voltage-Enhancement Mechanisms of an Organic Dye in High Open-Circuit Voltage Solid-State Dye-Sensitized Solar Cells. ACS Nano 2011, 5, (27) Neale, N. R.; Frank, A. J. Size and Shape Control of Nanocrystallites in Mesoporous TiO 2 Films. J. Mater. Chem. 2007, 17, (28) Zhu, K.; Kopidakis, N.; Neale, N. R.; van de Lagemaat, J.; Frank, A. J. Influence of Surface Area on Charge Transport and Recombination in Dye-Sensitized TiO 2 Solar Cells. J. Phys. Chem. B 2006, 110, (29) Zhao, Y.; Zhu, K. Charge Transport and Recombination in Perovskite (CH 3 NH 3 )PbI 3 Sensitized TiO 2 Solar Cells. J. Phys. Chem. Lett. 2013, 4, (30) Zhao, Y.; Zhu, K. Optical Bleaching of Perovskite (CH 3 NH 3 )- PbI 3 Through Room-Temperature Phase Transformation Induced by Ammonia. Chem. Commun. 2014, 50, (31) Dualeh, A.; Te treault, N.; Moehl, T.; Gao, P.; Nazeeruddin, M. K.; Graẗzel, M. Effect of Annealing Temperature on Film Morphology of Organic Inorganic Hybrid Pervoskite Solid-State Solar Cells. Adv. Funct. Mater. 2014, DOI: /adfm (32) Heo, J. H.; Im, S. H.; Noh, J. H.; Mandal, T. N.; Lim, C. S.; Chang, J. A.; Lee, Y. H.; Kim, H. J.; Sarkar, A.; Nazeeruddin, M. K.; et al. Efficient Inorganic-Organic Hybrid Heterojunction Solar Cells Containing Perovskite Compound and Polymeric Hole Conductors. Nat. Photonics 2013, 7, (33) Dualeh, A.; Moehl, T.; Te treault, N.; Teuscher, J.; Gao, P.; Nazeeruddin, M. K.; Graẗzel, M. Impedance Spectroscopic Analysis of Lead Iodide Perovskite-Sensitized Solid-State Solar Cells. ACS Nano 2014, 8, (34) Conings, B.; Baeten, L.; De Dobbelaere, C.; D Haen, J.; Manca, J.; Boyen, H.-G. Perovskite-Based Hybrid Solar Cells Exceeding 10% Efficiency with High Reproducibility Using a Thin Film Sandwich Approach. Adv. Mater. 2014, 26, dx.doi.org/ /jp502696w J. Phys. Chem. C 2014, 118,

8 Supporting Information CH 3 NH 3 Cl-Assisted One-Step Solution Growth of CH 3 NH 3 PbI 3 : Structure, Charge- Carrier Dynamics, and Photovoltaic Properties of Perovskite Solar Cells Yixin Zhao and Kai Zhu* Chemical and Materials Science Center, National Renewable Energy Laboratory, Denver West Parkway, Golden, Colorado (USA) Figure S1. UV-vis absorption spectra of perovskite CH 3 NH 3 PbI 3 on mesoporous TiO 2 films as a function of annealing time at 100 o C using precursors containing (a) 0 MACl, (b) 0.5 MACl, (c) 1 MACl, and (d) 2 MACl. 1

9 Table S1. Effect of MACl Amount (x) on Short-Circuit Photocurrent Density J sc, Open-Circuit Voltage V oc, Fill Factor FF, and Conversion Efficiency η of Solid-State Mesostructured (Meso) and Planar Perovskite CH 3 NH 3 PbI 3 Solar Cells. The mean values and standard deviations of the PV parameters from cells for each type of devices are given in parentheses. Cell Type (x) J sc (ma/cm 2 ) V oc (V) FF η (%) Meso (0) Meso (0.5) Meso (1) Meso (2) Planar (0) Planar (0.5) Planar (1) Planar (2) (16.96±0.64) (17.91±0.67) (19.44±0.61) (19.38±0.50) 5.55 (4.75±0.82) (17.38±0.66) (20.08±0.76) (19.84±0.63) (0.826±0.013) (0.836±0.016) (0.824±0.019) (0.823±0.016) (0.735±0.060) (0.974±0.022) (1.019±0.029) (1.013±0.042) (0.545±0.032) (0.552±0.039) (0.565±0.023) (0.597±0.020) (0.375±0.035) (0.561±0.038) (0.515±0.048) (0.540±0.029) 8.23 (7.64±0.64) 9.12 (8.25±0.54) 9.57 (9.03±0.33) (9.52±0.37) 1.86 (1.34±0.39) (9.50±0.85) (10.51±0.92) (10.85±0.79) Figure S2. Dark J V curves of (a) mesostructured and (b) planar perovskite solar cells as a function of added amount of MACl in the precursor solution for growing CH 3 NH 3 PbI 3. 2

10 Figure S3. Typical Nyquist plots of the impedance responses for a planar perovskite cell with three different bias voltages. The model used for impedance analysis has been previously discussed by others. 1-2 Figure S4. Typical large-scale SEM images of CH 3 NH 3 PbI 3 films grown on planar TiO 2 compact layer. 3

11 Figure S5. Effect of MACl on (a) electron diffusion coefficient as a function of photoelectron density and (b) recombination lifetime as function of voltage in mesostructured perovskite CH 3 NH 3 PbI 3 solar cells. Charge transport and recombination properties in mesostructured perovskite CH 3 NH 3 PbI 3 solar cells are studied by IMPS and IMVS as described previously. 3-4 Figure S5a shows the effect of using MACl on the diffusion coefficient (D) as a function of photoelectron density (n). All cells exhibit essentially the same power-law dependence (D n 1/α 1, with α being a disorder parameter) that is attributable to the electrons undergoing multiple trapping and detrapping through the mesoporous electrode film There is no obvious difference of the D values for mesostructured perovskite cells using different amounts of MACl in the CH 3 NH 3 PbI 3 precursor solution, suggesting that using MACl does not affect the trap distribution on the TiO 2 surface. Similarly, no significant difference is observed for the recombination lifetime as a function of voltage for mesostructured cells prepared using different amounts of MACl (Figure S5b). These results imply that the charge collection (determined by the competition between transport and recombination) is not affected by the use of MACl. 4

12 Figure S6. UV-vis absorption spectra of perovskite film as a function of annealing time at 100 o C using a precursor containing equimolar mixture of MACl and PbI 2. Figure S7. J V curves of planar perovskite solar cell based on the mixed halide CH 3 NH 3 PbI 3-x Cl x prepared from the precursor containing MAI and PbCl 2 (3:1 molar ratio). The cell efficiency is 11.86% with a J sc of ma/cm 2, V oc of V, and FF of REFERENCES (1). Juarez-Perez, E. J.; Wuβler, M.; Fabregat-Santiago, F.; Lakus-Wollny, K.; Mankel, E.; Mayer, T.; Jaegermann, W.; Mora-Sero, I. Role of the Selective Contacts in the Performance of Lead Halide Perovskite Solar Cells. J Phys. Chem. Lett. 2014, 5, (2). Christians, J. A.; Fung, R. C. M.; Kamat, P. V. An Inorganic Hole Conductor for Organo-Lead Halide Perovskite Solar Cells. Improved Hole Conductivity with Copper Iodide. J. Am. Chem. Soc. 2014, 136,

13 (3). Zhao, Y.; Zhu, K. Charge Transport and Recombination in Perovskite (CH 3 NH 3 )PbI 3 Sensitized TiO 2 Solar Cells. J. Phys. Chem. Lett. 2013, 4, (4). Zhao, Y.; Nardes, A. M.; Zhu, K. Solid-State Mesostructured Perovskite CH3NH3PbI3 Solar Cells: Charge Transport, Recombination, and Diffusion Length. J. Phys. Chem. Lett. 2014, 5, (5). Nelson, J.; Haque, S. A.; Klug, D. R.; Durrant, J. R. Trap-Limited Recombination in Dye- Sensitized Nanocrystalline Metal Oxide Electrodes. Phys. Rev. B 2001, 63, (6). Dloczik, L.; Ileperuma, O.; Lauermann, I.; Peter, L. M.; Ponomarev, E. A.; Redmond, G.; Shaw, N. J.; Uhlendorf, I. Dynamic Response of Dye-Sensitized Nanocrystalline Solar Cells: Characterization by Intensity-Modulated Photocurrent Spectroscopy. J. Phys. Chem. B 1997, 101, (7). Solbrand, A.; Lindstrom, H.; Rensmo, H.; Hagfeldt, A.; Lindquist, S. E.; Sodergren, S. Electron Transport in the Nanostructured TiO 2 -Electrolyte System Studied with Time-Resolved Photocurrents. J. Phys. Chem. B 1997, 101, (8). Bisquert, J.; Vikhrenko, V. S. Interpretation of the Time Constants Measured by Kinetic Techniques in Nanostructured Semiconductor Electrodes and Dye-Sensitized Solar Cells. J. Phys. Chem. B 2004, 108, (9). Barzykin, A. V.; Tachiya, M. Mechanism of Charge Recombination in Dye-Sensitized Nanocrystalline Semiconductors: Random Flight Model. J. Phys. Chem. B 2002, 106, (10). Zhu, K.; Kopidakis, N.; Neale, N. R.; van de Lagemaat, J.; Frank, A. J. Influence of Surface Area on Charge Transport and Recombination in Dye-Sensitized TiO 2 Solar Cells. J. Phys. Chem. B 2006, 110,