The Origin of Lower Hole Carrier Concentration in Methylammonium Tin Halide Films Grown by a Vapor-Assisted Solution Process Takamichi Yokoyama,†,‡ Tze-Bin Song,† Duyen H. Cao,† Constantinos C. Stoumpos,† Shinji Aramaki,‡ and Mercouri G. Kanatzidis*,† †
Department of Chemistry, Northwestern University, 2145 Sheridan Road, Evanston, Illinois 60208, United States Mitsubishi Chemical Group Science & Technology Research Center, Inc., 1000 Kamoshida-cho, Aoba-ku, Yokohama 227-8502, Japan
‡
S Supporting Information *
ABSTRACT: A low hole carrier concentration in methylammonium tin halide (MASnX3) perovskite semiconductors is a prerequisite for a nonshorting solar cell device. In-depth film characterizations were performed on MASnI3−xBrx films, fabricated by both a low-temperature vapor-assisted solution process (LTVASP) and conventional one-step methods, to reveal the origin of the lower hole carrier concentration from films of the former approach. We found that the vaporization of CH3NH3I solid at 150 °C, the temperature at which the LT-VASP occurs, does not supply iodine to the SnX2 (X = Br, I) films. As a result, secondary phases form aside from the desired MASnX3 perovskite; the secondary phases are suggested to be SnO and Sn(OH)2 via a proposed reaction pathway and are further supported by X-ray photoemission spectroscopy (XPS). These nonperovskite Sn2+ phases are beneficial because they assist in achieving the lower hole-doping levels in LT-VASP films. Remarkably, LT-VASP devices demonstrate improved air stability. Overall, our findings suggest that not only the commonly used SnF2 but also other divalent Sn compounds could serve as Sn vacancy suppressors. Further work on modulating the perovskite film compositions could realize more efficient and stable tin-based perovskite solar cells. from the easy oxidation of Sn2+ to Sn4+.19−21 Hence, one could certainly fail to obtain functional ASnX3-based solar cells without appropriate modification of the perovskite active layer.3,4,17 Previous works have introduced SnF2 into CsSnI3 to suppress the Sn2+ vacancies,9,22 and this strategy has been applied to other Sn-based perovskite solar cell systems, depicting encouraging results. Recently, Lee et al. communicated the fabrication of FASnI3 solar cells using a SnF2-pyrazine complex, achieving a promising PCE of ∼4.8% with high reproducibility.23 We have demonstrated the successful fabrication of MASnI3based devices using a low-temperature vapor-assisted solution process (LT-VASP); in this process, MASnI3 film was formed upon the reaction of a SnI2 film (solid) and MAI (gas).24 In
R
emarkable progress has been made in developing highly efficient solid-state solar cells incorporating organic− inorganic lead halide perovskites as light-absorbing materials with certified power conversion efficiency (PCE) surpassing 20%.1,2 However, the toxicity of the lead element from the archetypal methylammonium (MA) lead iodide (MAPbI3) perovskite may well hinder the large-scale deployment of this technology. It is therefore desirable to substitute Pb with less toxic metals. A good variety of lead-free perovskite compounds has been utilized for solar cell application, including but not limited to MA tin halide (MASnI3−xBrx (x = 0−3)), formamidinium tin halide (FASnI 3−x Br x ), CsSnI 3−x Br x , MA(or Cs)GeI 3 , Cs 3 Sb 2 I 9 , and MA(or Cs)3Bi2I9.3−18 So far, the divalent Sn-based perovskites outperform all of the currently reported lead-free materials in terms of PCE. Sn-based perovskites, despite being regarded as semiconductors, often display metallic-like behavior due to the inadvertent and/or spontaneous hole carrier doping resulting © XXXX American Chemical Society
Received: October 7, 2016 Accepted: November 28, 2016
22
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
Letter
http://pubs.acs.org/journal/aelccp
Letter
ACS Energy Letters comparison with a film fabricated by the conventional one-step deposition approach, one obvious major advantage of the LTVASP MASnI3 film is the excellent film uniformity and surface coverage. In addition, our Kelvin Probe data suggested that the LT-VASP MASnI3 film had a lower p-doping level (or lower dark background hole carrier density). Assuming that both the one-step and VASP methods yield the same MASnI3 film composition, the question arises as to what is the origin of this difference. The lower hole carrier concentration is strongly believed to be responsible for suppressing the metallic characteristics of as-deposited MASnI3 films and the resulting working solar cells. To better understand the origin of the lower hole-doping level in the LT-VASP MASnI3 films and its effects in photovoltaic performance, we set out to study the properties of MASnI3−xBrx films prepared by both LT-VASP and one-step approaches. The compositional difference between these two approaches is best illustrated in the mixed halide MASnI3−xBrx systems. To our surprise, the vaporization of MAI powder at LT-VASP temperature conditions (150 °C) does not supply iodine to SnX2 films and essentially yields secondary phases, SnO and Sn(OH)2, in the final MASnI3 films. Additionally, we show that the secondary Sn2+ phases have a strong impact on the air stability of the Sn-based perovskite devices. In our previous report, the one-step MASnI3 film was observed to have poorer surface coverage than the LT-VASP film.24 Although our Kelvin Probe and device studies have suggested that the one-step films have a higher p-doping level than the LT-VASP films resulting in poorer device performance, the effect of film morphology and surface coverage on device performance also plays an important role. Therefore, we started out by investigating the properties and device performance of one-step MASnI3 films because of the better coverage. The good-quality films were fabricated by mixed solvent of N,N-dimethylformamide and dimethyl sulfoxide (DMF/DMSO) or the antisolvent dripping technique, as shown in Figure S1a.5,25,26 Four-point probe resistance measurements were employed to characterize the conductivity (p-doping levels) of the thin films. Although the thickness of the perovskite capping layer on TiO2 substrates may affect the measurement, it was obvious that the one-step films have significantly lower resistance than the LT-VASP films, as shown in Figure S1b and Table S1. Moreover, these one-step MASnI3based devices showed short-circuit behavior despite the higher surface coverage films (Figure S2). This short-circuit response is similar to the performance of MASnI3 devices prepared by the one-step method using DMF, as reported in our previous study.24 These results further support that LT-VASP films have lower p-doping level and motivate us to further investigate the compositional difference between the one-step and VASP films. Our MASnI3−xBrx films were deposited on a mesoporous TiO2 substrate (350 nm) by both conventional one-step and LT-VASP methods. In the LT-VASP, MAI (gas) was reacted with films of SnI2/SnBr2 (solid) or SnBr2 (solid). For convenience, we refer to these two films as MAI:SnI2/SnBr2 and MAI/SnBr2 films, respectively. To probe the crystallinity and optical absorption properties of the thin films, X-ray diffraction (XRD) and ultraviolet−visible (UV−vis) spectroscopy were used. XRD patterns of MASnI3−xBrx films are shown in Figure 1a. With increased Br content, 2θ angles of Bragg peaks shift to higher values for both one-step and LT-VASP films, indicating the successful solid substitution of Br into the MASnI3 crystal structure.3 Notably, given the expected
Figure 1. (a) XRD patterns and (b) Tauc plot of one-step and LTVASP MASnI3−xBrx(x = 0−3) films. In (a), a simulated MASnI3 pattern is also shown for comparison.* indicates peaks from the Fdoped tin oxide (FTO)/TiO2 substrate. In (b), (αhν)2 is estimated from absorption spectra and plotted as a function of energy (i.e., Tauc plot for the direct band gap. α, h, and ν indicate absorbance, the Planck constant, and the wavenumber, respectively).
perovskite product of the LT-VASP MAI/SnI2/SnBr2 film to be MASnI2Br (x = 1), its XRD pattern instead shows the 2θ values lying between those of the one-step MASnI2Br and MASnIBr2 films, offering the first hint of the imperfect reaction between MAI (gas) and SnI2/SnBr2 (solid). To shed light on our initial finding, we further examined the LT-VASP MAI/ SnBr2 film. The XRD peaks of this “nominal” MASnIBr2 (x = 2) film match well with those of the one-step MASnBr3, illustrating that the evaporation of MAI powder at 150 °C does not supply any iodine to the SnBr2 films. A more pronounced discrepancy between the one-step and the LT-VASP films is illustrated in their UV−vis spectra and Tauc plots of them, as shown in Figures S3 and 1b, respectively. We summarize the band gap (Eg) of each film estimated from the Tauc plot in Table 1.27 In agreement with the observations from XRD Table 1. Band Gaps of One-Step and LT-VASP MASnI3−xBrx Filmsa one-step films [or powder (ref 3)]
LT-VASP films
material
band gap (eV)
material
band gap (eV)
MASnI3 MASnI2Br MASnIBr2 MASnBr3
1.35 (1.30) 1.54(1.56) 1.81 (1.75) 2.18(2.15)
MAI(g)/SnI2 MAI(g)/SnI2/SnBr2 MAI(g)/SnBr2
1.25 1.67 2.13
a
Values in parentheses are the band gaps of powder samples reported in the literature.3
23
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
Letter
ACS Energy Letters
Figure 2. XPS spectra of MASnI3 films. (a) Survey scans of the one-step and LT-VASP MASnI3 films. (b) O 1s peaks of the LT-VASP film. (c) O 1s peak of the LT-VASP film before and after sputtering for 15 and 75 s. (d) Sn 3d5/2 peak of the LT-VASP film.
films with X-ray photoelectron spectroscopy (XPS). All of the elements in the MASnI3 films were identified as shown in Figure 2a. A significant difference in the Sn/I ratio was found, which is consistent with the EDS results. A more significant and important difference between the two films comes from the O 1s peak, which was only detected from the LT-VASP films, as shown in Figures 2a,b and S6. We confirmed that the O peak was not due to oxygen chemically or physically adsorbed on the surface by repeatedly detecting it from films that have undergone multiple sputtering cycles using an Ar-gun source (Figure 2c). Moreover, the O 1s peak shows at least two states consistently in the depth profile scan, suggesting two different oxidation states/coordination environments. We attribute these two peaks to SnO and Sn(OH)2 phases present in the films, whose binding energies are consistent with the reported values.28,29 The Sn 3d peak in Figure 2d also shows a subtle shoulder, indicating the presence of at least two different oxidation Sn states in which the lower and higher binding energies are assigned to Sn2+ and Sn4+, respectively. The proposed Sn4+ state may result from the brief air exposure of our films for approximately 30 s before loading into the XPS vacuum chamber. Alternatively, the subtle shoulder may reflect upon the different coordination around the Sn metal, with the difference arising from the different electronegativity of oxygen and iodide. Because the binding energies of Sn2+ in MASnI3, SnO, and Sn(OH)2 are so close to one another, we could not deconvolute the Sn2+ peak into these three peaks accurately. Nonetheless, the XPS results reveal that the LT-VASP yields MASnI3 with two different oxide impurities. Knowing that evaporating MAI during LT-VASP does not supply iodine into the Sn halide films, we propose a reaction pathway that leads to the formation of the composite films containing not only the MASnI3 perovskite but also other oxide compounds as secondary phases. At the elevated temperature (150 °C) conditions of LT-VASP, the starting reactant MAI powder will partially equilibrate into methylamine gas MA(g) and HI(g).30 HI(g) may then decompose further into I2 and H2O in the presence of O2;31 a very low concentration of O2 is
patterns, the optical absorption spectrum of the LT-VASP MAI/SnI2/SnBr2 film lies between those of one-step MASnI2Br (Eg = 1.54 eV) and MASnIBr2 (Eg = 1.81 eV). Along the same line, the LT-VASP MAI/SnBr2 film shows significantly larger Eg than its expected nominal composition MASnIBr2 (Eg = 1.54 eV). In fact, the absorption spectrum of the MAI/SnBr2 film almost coincides with that of the one-step MASnBr3. The slight difference between the one-step and LT-VASP MASnI3 films is attributed to the slight oxidization of the one-step films during the film formation process and during the UV−vis measurement because they are more air sensitive than the LT-VASP films.24 Photographs of the two films can be found in the Supporting Information (Figure S4), showing that the LTVASP MAI/SnBr2 film has the same orange color as the onestep MASnBr3 film. It is clear from the XRD patterns and UV− vis spectra that the MAI (gas)/SnX2 (solid) reaction during LT-VASP is imperfect as the evaporation of MAI powder at 150 °C supplies MA and/or other gases but not MAI gas to the SnX2 films, producing “non-nominal” MASnX3 films. Energy dispersive X-ray spectroscopy (EDS) was performed to quantitatively determine the compositions of the LT-VASP MASnI3−xBrx films with more details in Figure S5. The I/Br ratio in MAI/SnI2/SnBr2 films was estimated to be 1:1, while iodine was not detected at all in MAI/SnBr2 films; the actual compositions were found to be MASnI1.5Br1.5 and MASnBr3, respectively. In other words, the I/Br ratio stays constant before and after LT-VASP, and it is determined by the ratio of SnI2/ SnBr2 alone. Additionally, the Sn/I ratio of our LT-VASP MASnI3 films, grown on a glass substrate to avoid the Sn background from the FTO substrate, was also confirmed to be indeed 1:2 as it is in SnI2. Given that the Sn concentration is richer than what it should be, it is reasonable to presume that the LT-VASP MASnI3−xBrx films possess some additional Snbased secondary phases, making them “Sn-rich” films, and this likely leads to the difference in the hole-doping levels between the LT-VASP and one-step films. To further understand the formation mechanism of the Snrich films, we characterized the one-step and LT-VASP MASnI3 24
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
Letter
ACS Energy Letters
Finally, we report the preliminary results of the photovoltaic performance of our MASnI3−xBrx-based devices prepared by both LT-VASP and one-step methods. The devices were composed of six layers, including FTO, TiO2 (compact/ mesoporous layers), perovskite, a hole transport layer (HTL), and the Au top electrode. Current density−voltage (J−V) characteristics and device parameters are shown in Figure 3a
expected to present in our glovebox atmosphere. Figure S7 shows the after LT-VASP MAI powder, which has turned yellow, further confirming the I2 byproduct. Because the formation energy of MASnI3 is low,32 which is clearly proven by its spontaneous formation at room temperature,3,4 a trace amount of H2O, which may originate from the hygroscopic MAI powder, our glovebox atmosphere, and decomposed products of HI, would be enough to drive the conversion of SnI2 into MASnI3. Equations 1−3 highlight the possible reaction mechanism,33 in which the stoichiometric MASnI3 is formed along with the Sn surplus products, SnO and Sn(OH)2. 3SnI 2 + 2CH3NH 2 + H 2O → 2CH3NH3SnI3 + SnO (1)
CH3NH 2 + H 2O ↔ CH3NH3OH
(2)
3SnI 2 + 2CH3NH3OH → 2CH3NH3SnI3 + Sn(OH)2 (3)
To rationalize our hypothesis, we proceeded to use methylammine gas (MA(g)), an iodine-free reactant, instead of MAI powder to react with SnX2 films, based on a recent report that employed a similar approach to prepare MAPbI3 perovskite films.33 Upon exposure of the SnX2 film to the vapor from an ethanol solution of MA at room temperature, SnI2, SnI2/SnBr2, and SnBr2 thin films turned black, brown, and orange, respectively, within a few minutes. In this process, H2O from the ethanol solution of methylamine may also accelerate the reaction. The obtained films had an extremely smooth and shiny surface. More importantly, their XRD and UV−vis spectra are similar to those of the LT-VASP films (Figure S8a,b), validating our idea that only the MA part of MAI powder reacted with the SnX2 films. The reaction mechanism operating during LT-VASP might in reality be more complex than our proposed pathway. Yet, LT-VASP MASnI3 films have been proven to contain excess Sn2+ instead of the ideal 1:3 Sn/halide ratio from our XRD, optical absorption, EDS, and XPS results. Theoretical and experimental studies have reported that excess Sn2+ can suppress Sn vacancies in the Sn-based perovskite films, thus lowering the p-doping level.9,34 So far, SnF2 has been used as the excess Sn2+ source; our studies suggest that other Sn2+ sources, such as SnO or Sn(OH)2, could also work for compensating Sn vacancies in a similar manner to SnF2. Additionally, it is important to mention that the excess amount of Sn2+ phases in the LT-VASP film is quite significant. Specifically, the Sn/I ratio is 1:2 based on our EDS results, which means that we have a 50% excess of Sn2+ (relative to Sn2+ in MASnI3), assuming that no oxidation of Sn2+ from MASnI3 occurs. This amount is much higher than the SnF2 amount introduced in the other Sn-based perovskite systems in the literature (∼20%). Given that the addition of over 20% SnF2 deteriorates the device performance, mainly due to poor film growth,9,23 it has not been shown whether the large excess of Sn2+, in general, is actually harmful. The spatial distribution of this significant amount of the secondary phase is still under investigation. However, considering the fact that the LT-VASP MASnI3 films did not show any extra Bragg peaks in the XRD patterns or agglomeration in SEM images,24 it is plausible to consider that the SnO and Sn(OH)2 disperse uniformly into the MASnI3 matrix. Our present results suggest the unique nature of Sn-based perovskites in which more than 50% excess Sn2+ is still beneficial for the device performance unless the morphology of the perovskite films is deteriorated.
Figure 3. (a) J−V characteristics and (b) IPCE spectra of one-step and LT-VASP MASnI3−xBrx-based devices. Only in the case of LTVASP MAI/SnI2/SnBr2 were different samples used for J−V and IPCE measurements. IPCEs of one-step devices except MASnBr3 were not measured because of their short-circuit behavior. A J−V curve and IPCE spectrum of LT-VASP MASnI3-based devices in our previous work are also shown for comparison.
Table 2. Summary of Photovoltaic Performanceb
MAI(g) + SnI2 (ref 24) MAI(g) + SnI2/SnBr2 MAI(g) + SnBr2 one-step MASnBr3
JSC (mA/cm2)
VOC (V)
fill factor (%)
PCE (%)
17.36 5.02 2.24 1.22
0.273 0.452 0.487 0.307
39.1 48.3 46.4 36.8
1.86 1.10 0.51 0.14
b
One-step MASnI2Br-based and MASnIBr2-based devices were shortcircuited and are thus not presented in this table.
and Table 2. While the MASnI2Br and MASnIBr2 devices fabricated from the one-step method showed short-circuit behavior, all devices fabricated from the LT-VASP showed functional diode characteristics, proving once again the importance of the low hole-doping levels. The MAI/SnI2/ SnBr2 and MAI/SnBr2-based devices (namely, LT-VASP MASnI1.5Br1.5 and LT-VASP MASnBr3) showed higher opencircuit voltage (VOC) but lower short-circuit current (JSC) than the LT-VASP MASnI3 device, as expected from the Eg reduction from Br to I. Note that the LT-VASP MASnBr3 25
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
Letter
ACS Energy Letters films have relatively poor surface coverage, as shown in Figure S9; thus, with better film quality, a higher VOC for the MASnBr3 device is entirely achievable even though the limitation would still be determined by the conduction band maximum of TiO2.24 Figure 3b shows the typical incident photon-to-current conversion efficiency (IPCE) excluding short-circuited devices. The IPCE became lower in the longer-wavelength region in every device, which is consistent with previous results in various Sn-based devices and Sn-rich devices in Sn−Pb mixed systems.6,7,9,35 This could be due to weaker absorbance in the longer-wavelength region, particularly when x in MASnI3−xBrx is low, as shown in Figure S3. Weaker absorbance allows the incident light to pass through deeper in the light absorber. This leads to carrier generation far away from the electron transport layer (ETL) interface, and thus, the photogenerated electrons need to travel a longer distance to be extracted. Also, MASnI3−xBrx carriers have a shorter electron diffusion length owning to their background hole carrier density. These two factors reduce the survival time of the photogenerated electrons, resulting in lower internal quantum efficiency at the longer wavelengths. The opposite feature (namely, the stronger absorbance region shows lower IPCE) should be observed in an inverted device structure where the perovskite layer is formed on the HTL instead of the ETL, which in fact has been reported recently.36 The IPCE onsets correspond well with the optical Eg. We attribute the relatively low IPCE (thus, JSC) of LT-VASP MASnI1.5Br1.5 and LT-VASP MASnBr3 devices to the not-yet-optimized film quality because the LTVASP conditions used in this study were optimized for the MASnI3 member only. Because our main focus of this study is to understand the difference between the one-step and LT-VASP films, we chose the MASnBr3-based devices for further study because they were the only system that shows functional devices for both film fabrication methods. The IPCE spectra from two devices have similar shape, confirming that the two films are nominally the same perovskite material. The LT-VASP film is however remarkably more robust to air exposure than the one-step film. Figure 4 shows J−V characteristics measured in air without encapsulation. The performance of the one-step MASnBr3 devices decreased significantly due to leakage current, as highlighted by the current increase in the reverse bias region. Furthermore, the fill factor dropped significantly due to the “Sshaped” J−V characteristic. This behavior could stem from a strong energy barrier at the perovskite/HTL interface or
imbalance of carrier transport, probably because of the formation of the oxidation product MA2SnBr6 at the interface of the perovskite and HTM layer.37 As a result, the device retained only 3% of the original performance after being exposed to air for 30 min. In contrast, the LT-VASP MASnBr3 devices had much better air stability. Although slight degradation still occurred, the device retained 79 and 69% of the original performance after exposing to air for 30 and 60 min, respectively. The results indicate that the excess Sn2+containing phases from LT-VASP films are beneficial not only to suppressing the short-circuit behavior but also to enhancing the device stability in ambient conditions. In conclusion, the evaporation of CH3NH3I powder during the LT-VASP film growth of CH3NH3SnX3 only supplies CH3NH2 gas to react with the SnI2 films. As a result, the final LT-VASP CH3NH3SnI3 films contain not only CH3NH3SnI3 but also SnO and Sn(OH)2, making them Sn2+-rich in nature. We believe that the excess Sn2+ compounds are responsible for the advantageous low hole-doping level in the LT-VASP MASnX3 films, suppressing oxidation and preventing the devices from short-circuit behavior. Our study suggests that not only SnF2 but also other divalent Sn sources such as SnO and Sn(OH)2 can be potentially used to regulate the Sn vacancies in the perovskite crystal structure. More importantly, our results illustrate the complex nature of Sn-based perovskites where the 50% excess of Sn2+ second phases does not degrade but indeed improves the device performance. Additionally, the Sn2+ excess is beneficial to the device’s air stability. Finally, our study indicates that a more sophisticated composition modulation approach, including the secondary “spectator” phases present in the films with proper device optimization, could realize more efficient and stable Sn-based perovskite solar cells.
■
METHODS SnI2 synthesis and purification, FTO/TiO2 (compact layer)/ TiO2 (mesoporous layer) substrate preparation, LT-VASP, hole transport material (HTM) preparation, device fabrication (HTM coating and Au evaporation), and IPCE measurements were performed according to our previous report.24 Materials. CH3NH3I and CH3NH3Br were purchased from Lumtec. CH3NH2 solution (33 wt % in absolute ethanol) and poly[bis(4-phenyl)(2,4,6-trimethylphenyl)amine] (PTAA) were purchased from Sigma-Aldrich. SnBr2 (99+%) and 4isopropyl-4′-methyldiphenyliodonium tetrakis(pentafluorophenyl)borate (TPFB) were purchased from Alfa Aesar and TCI, respectively. All materials were used as received. Perovskite Film Fabrication by Gas−Solid Reaction (LT-VASP). MAI(g)/SnI2/SnBr2 films and MAI(g)/SnIBr2 films were deposited using the following procedure. SnI2 (0.5 mmol) and SnBr2 (0.5 mmol) together and SnBr2 (1 mmol) alone were dissolved in 1 mL of anhydrous N,N-dimethylformamide (DMF) to make a 1 M solution of SnI2/SnBr2 and SnBr2, respectively. The tin halide solutions were stirred at 70 °C for 1 h and filtered through a syringe filter with a 0.45 μm GHP membrane before use. A 45 μL SnI2/SnBr2 (or SnBr2) solution was dropped onto the mesoporous TiO2 substrate, and the substrate was spun at 2000 rpm for 60 s to form the SnI2/SnBr2 (or SnBr2) films. LT-VASP was performed for these films to form perovskite films. All of these procedures were performed in a N2-filled glovebox. When MA(g) was used for the reaction, each tin halide film was exposed to MA solution for a few seconds. The film color changed to the corresponding
Figure 4. J−V characteristics of one-step and LT-VASP MASnBr3based devices measured in air without encapsulation. The performance degraded more significantly for one-step-based devices. 26
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
ACS Energy Letters
■
perovskite colors at room temperature after 1 min. Subsequently, the films were annealed at 70 °C for 10 min to remove residual solvents. Perovskite Film Fabrication by the One-Step Method. 1 M of MASnI3, MASnI2Br, MASnIBr2, and MASnBr3 in DMF solutions were prepared from MAI (1 mmol) and SnI2 (1 mmol), MAI (1 mmol), SnI2 (0.5 mmol), and SnBr2 (0.5 mmol), MAI (1 mmol) and SnBr2 (1 mmol), and MABr (1 mol) and SnBr2 (1 mmol), respectively. The solutions were stirred at 70 °C for 1 h and filtered before use. Then, 45 μL of each solution was dropped onto the mesoporous TiO2 substrate, and the substrate was spun at 2000 rpm for 60 s to form methylammonium tin halide perovskite films. The MASnI3 films used for the film resistance study were formed on glass/TiO2 (compact layer)/TiO2 (mesoporous layer) substrates from MASnI3 in DMF/DMSO (4:1) mixed solvent for better coverage. For some films, diethyl ether as an antisolvent was dripped during the spin-coating process according to the literature.26 All of the films were then annealed at 70 °C for 10 min after the spin-coating. The higher surface coverage MASnI3 films formed in the DMF/DMSO mixed solvent were also used for XPS study to prevent a TiO2 signal. All of these procedures were performed in the N2-filled glovebox. Perovskite Film Characterization. The resistance of the MASnI3 films was measured in the N2-filled glovebox by the four-point probes method using a Keithley 2401 source meter where the distance between probes was 1.8 cm. Transmittance measurements were performed from 300 to 1200 nm at room temperature using a Shimadzu UV-3600 PC double-beam, double-monochromator spectrophotometer. UV−vis absorption spectra were calculated from transmittance by using an equation of α = −log(T), where α and T are the absorbance and transmittance, respectively. XRD was collected using a Rigaku MiniFlex600 X-ray diffractometer (Cu Kα, 1.5406 Å) operating at 40 kV and 20 mA. Either a Hitachi 4800 or 8030 SEM was used for surface morphology and EDS studies. XPS measurements were carried out on an Omicron ESCA Probe XPS spectrometer (Thermo Scientific ESCALAB 250Xi) using a 150 eV pass energy and 1 eV step size for the survey scan and a 20 eV pass energy and 0.01 eV step size for the fine scan. The depth profile was obtained by Ar-ion etching with 1 keV accelerating for 15 s per cycle. Solar Cell Characterization. Unless stated otherwise, all devices were encapsulated using a 30 μm thick hot-melting polymer and a microscope coverslip in the N2-filled glovebox. J−V measurements were carried out in air under 1 sun illumination (AM 1.5G, 100 mW/cm2) using a certified solar simulator (Abet Technologies) and Keithley 2400 source meter. The voltage was swept from positive bias to negative bias (namely, from VOC to JSC). To investigate the air stability, one-step and LT-VASP MASnBr3 devices were measured without encapsulation in air. J−V characteristics were measured right after the devices were exposed to air and after exposing them to air for 30 min. During the exposure, 1 sun light continued to irradiate the devices. The device active area was between 0.08 and 0.15 cm2, defined by the overlapping area between the unetched FTO glass and Au electrode.
Letter
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsenergylett.6b00513. Resistance measurements and device results of MASnI3 films, photographs and UV−vis of the MASnI3−xBrx films, SEM/EDS images and elemental analyses, and additional XRD, UV−vis, and XPS plots (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. ORCID
Takamichi Yokoyama: 0000-0002-0766-4550 Constantinos C. Stoumpos: 0000-0001-8396-9578 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS T.-B.S. acknowledges financial support from Mitsubishi Chemical Group Science & Technology Research Center, Inc. D.H.C. acknowledges support from the Link Foundation through the Link Foundation Energy Fellowship Program. This work was supported in part by the ANSER Center, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, and Office of Basic Energy Sciences under Award Number DE-SC0001059. This work made use of the EPIC facility (NUANCE Center-Northwestern University), which has received support from the MRSEC program (NSF DMR-1121262) at the Materials Research Center, and the Nanoscale Science and Engineering Center (EEC-0118025/ 003), both programs of the National Science Foundation; the State of Illinois; and Northwestern University.
■
REFERENCES
(1) Jeon, N. J.; Noh, J. H.; Yang, W. S.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. Compositional Engineering of Perovskite Materials for High-Performance Solar Cells. Nature 2015, 517, 476−480. (2) 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. (3) Hao, F.; Stoumpos, C. C.; Cao, D. H.; Chang, R. P.; Kanatzidis, M. G. Lead-Free Solid-State Organic-Inorganic Halide Perovskite Solar Cells. Nat. Photonics 2014, 8, 489−494. (4) Noel, N. K.; Stranks, S. D.; Abate, A.; Wehrenfennig, C.; Guarnera, S.; Haghighirad, A.-A.; Sadhanala, A.; Eperon, G. E.; Pathak, S. K.; Johnston, M. B.; et al. Lead-Free Organic-Inorganic Tin Halide Perovskites for Photovoltaic Applications. Energy Environ. Sci. 2014, 7, 3061−3068. (5) Hao, F.; Stoumpos, C. C.; Guo, P.; Zhou, N.; Marks, T. J.; Chang, R. P.; Kanatzidis, M. G. Solvent-Mediated Crystallization of CH3NH3SnI3 Films for Heterojunction Depleted Perovskite Solar Cells. J. Am. Chem. Soc. 2015, 137, 11445−11452. (6) Sabba, D.; Mulmudi, H. K.; Prabhakar, R. R.; Krishnamoorthy, T.; Baikie, T.; Boix, P. P.; Mhaisalkar, S. G.; Mathews, N. Impact of Anionic Br− Substitution on Open Circuit Voltage in Lead Free Perovskite (CsSnI3−xBrx) Solar Cells. J. Phys. Chem. C 2015, 119, 1763−1767. (7) Koh, T. M.; Krishnamoorthy, T.; Yantara, N.; Shi, C.; Leong, W. L.; Boix, P. P.; Grimsdale, A. C.; Mhaisalkar, S. G.; Mathews, N. Formamidinium Tin-Based Perovskite with Low Eg for Photovoltaic Applications. J. Mater. Chem. A 2015, 3, 14996−15000. 27
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
Letter
ACS Energy Letters (8) Krishnamoorthy, T.; Ding, H.; Yan, C.; Leong, W. L.; Baikie, T.; Zhang, L.; Sherburne, M.; Li, S.; Asta, M.; Mathews, N.; Mhaisalkar, S. G. Lead-Free Germanium Iodide Perovskite Materials for Photovoltaic Application. J. Mater. Chem. A 2015, 3, 23829−23832. (9) Kumar, M. H.; Dharani, S.; Leong, W. L.; Boix, P. P.; Prabhakar, R. R.; Baikie, T.; Shi, C.; Ding, H.; Ramesh, R.; Asta, M.; et al. LeadFree Halide Perovskite Solar Cells with High Photocurrents Realized through Vacancy Modulation. Adv. Mater. 2014, 26, 7122−7127. (10) Park, B. W.; Philippe, B.; Zhang, X.; Rensmo, H.; Boschloo, G.; Johansson, E. M. Bismuth Based Hybrid Perovskites A3Bi2I9 (A: Methylammonium or Cesium) for Solar Cell Application. Adv. Mater. 2015, 27, 6806−6813. (11) Lyu, M.; Yun, J.-H.; Cai, M.; Jiao, Y.; Bernhardt, P. V.; Zhang, M.; Wang, Q.; Du, A.; Wang, H.; Liu, G.; Wang, L. Organic-Inorganic Bismuth (III)-Based Material: A Lead-Free, Air-Stable and SolutionProcessable Light-Absorber Beyond Organolead Perovskites. Nano Res. 2016, 9, 692−702. (12) Ö z, S.; Hebig, J.-C.; Jung, E.; Singh, T.; Lepcha, A.; Olthof, S.; Jan, F.; Gao, Y.; German, R.; van Loosdrecht, P. H.; et al. ZeroDimensional (CH3NH3)3Bi2I9 Perovskite for Optoelectronic Applications. Sol. Energy Mater. Sol. Cells 2016, 158, 195−201. (13) Jung, M.-C.; Raga, S. R.; Qi, Y. Properties and Solar Cell Applications of Pb-Free Perovskite Films Formed by Vapor Deposition. RSC Adv. 2016, 6, 2819−2825. (14) Marshall, K. P.; Walton, R. I.; Hatton, R. A. Tin Perovskite/ Fullerene Planar Layer Photovoltaics: Improving the Efficiency and Stability of Lead-Free Devices. J. Mater. Chem. A 2015, 3, 11631− 11640. (15) Zhang, M.; Lyu, M.; Yun, J.-H.; Noori, M.; Zhou, X.; Cooling, N. A.; Wang, Q.; Yu, H.; Dastoor, P. C.; Wang, L. Low-Temperature Processed Solar Cells with Formamidinium Tin Halide Perovskite/ Fullerene Heterojunctions. Nano Res. 2016, 9, 1570−1577. (16) Chen, Z.; Wang, J. J.; Ren, Y.; Yu, C.; Shum, K. Schottky Solar Cells Based on CsSnI3 Thin-Films. Appl. Phys. Lett. 2012, 101, 093901. (17) Ogomi, Y.; Morita, A.; Tsukamoto, S.; Saitho, T.; Fujikawa, N.; Shen, Q.; Toyoda, T.; Yoshino, K.; Pandey, S. S.; Ma, T.; et al. Ch3NH3SnxPb(1‑x)I3 Perovskite Solar Cells Covering up to 1060 nm. J. Phys. Chem. Lett. 2014, 5, 1004−1011. (18) Saparov, B.; Hong, F.; Sun, J.-P.; Duan, H.-S.; Meng, W.; Cameron, S.; Hill, I. G.; Yan, Y.; Mitzi, D. B. Thin-Film Preparation and Characterization of Cs3Sb2I9: A Lead-Free Layered Perovskite Semiconductor. Chem. Mater. 2015, 27, 5622−5632. (19) Takahashi, Y.; Obara, R.; Lin, Z.-Z.; Takahashi, Y.; Naito, T.; Inabe, T.; Ishibashi, S.; Terakura, K. Charge-Transport in Tin-Iodide Perovskite CH3NH3SnI3: Origin of High Conductivity. Dalton Trans. 2011, 40, 5563−5568. (20) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Semiconducting Tin and Lead Iodide Perovskites with Organic Cations: Phase Transitions, High Mobilities, and Near-Infrared Photoluminescent Properties. Inorg. Chem. 2013, 52, 9019−9038. (21) Stoumpos, C. C.; Kanatzidis, M. G. The Renaissance of Halide Perovskites and Their Evolution as Emerging Semiconductors. Acc. Chem. Res. 2015, 48, 2791−2802. (22) Chung, I.; Lee, B.; He, J.; Chang, R. P.; Kanatzidis, M. G. AllSolid-State Dye-Sensitized Solar Cells with High Efficiency. Nature 2012, 485, 486−489. (23) Lee, S. J.; Shin, S. S.; Kim, Y. C.; Kim, D.; Ahn, T. K.; Noh, J. H.; Seo, J.; Seok, S. I. Fabrication of Efficient Formamidinium Tin Iodide Perovskite Solar Cells through SnF2-Pyrazine Complex. J. Am. Chem. Soc. 2016, 138, 3974−3977. (24) Yokoyama, T.; Cao, D. H.; Stoumpos, C. C.; Song, T.-B.; Sato, Y.; Aramaki, S.; Kanatzidis, M. G. Overcoming Short-Circuit in LeadFree CH3NH3SnI3 Perovskite Solar Cells Via Kinetically Controlled Gas-Solid Reaction Film Fabrication Process. J. Phys. Chem. Lett. 2016, 7, 776−782. (25) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. I. Solvent Engineering for High-Performance Inorganic-Organic Hybrid Perovskite Solar Cells. Nat. Mater. 2014, 13, 897−903.
(26) Ke, W.; Stoumpos, C. C.; Logsdon, J. L.; Wasielewski, M. R.; Yan, Y.; Fang, G.; Kanatzidis, M. G. TiO2-ZnS Cascade Electron Transport Layer for Efficient Formamidinium Tin Iodide Perovskite Solar Cells. J. Am. Chem. Soc. 2016, 138, 14998−15003. (27) Tauc, J.; Grigorovici, R.; Vancu, A. Optical Properties and Electronic Structure of Amorphous Germanium. Phys. Status Solidi B 1966, 15, 627−637. (28) Hu, D.; Han, B.; Deng, S.; Feng, Z.; Wang, Y.; Popovic, J.; Nuskol, M.; Wang, Y.; Djerdj, I. Novel Mixed Phase SnO2 Nanorods Assembled with SnO2 Nanocrystals for Enhancing Gas-Sensing Performance toward Isopropanol Gas. J. Phys. Chem. C 2014, 118, 9832−9840. (29) Rogé, V.; Bahlawane, N.; Lamblin, G.; Fechete, I.; Garin, F.; Dinia, A.; Lenoble, D. Improvement of the Photocatalytic Degradation Property of Atomic Layer Deposited ZnO Thin Films: The Interplay between Film Properties and Functional Performances. J. Mater. Chem. A 2015, 3, 11453−11461. (30) Jain, S. M.; Philippe, B.; Johansson, E. M.; Park, B.-w.; Rensmo, H.; Edvinsson, T.; Boschloo, G. Vapor Phase Conversion of PbI2 to CH 3 NH 3 PbI 3 : Spectroscopic Evidence for Formation of an Intermediate Phase. J. Mater. Chem. A 2016, 4, 2630−2642. (31) Niu, G.; Li, W.; Meng, F.; Wang, L.; Dong, H.; Qiu, Y. Study on the Stability of CH3NH3PbI3 Films and the Effect of Post-Modification by Aluminum Oxide in All-Solid-State Hybrid Solar Cells. J. Mater. Chem. A 2014, 2, 705−710. (32) Glasser, L. Solid-State Energetics and Electrostatics: Madelung Constants and Madelung Energies. Inorg. Chem. 2012, 51, 2420−2424. (33) Raga, S. R.; Ono, L. K.; Qi, Y. Rapid Perovskite Formation by CH3NH2 Gas-Induced Intercalation and Reaction of PbI2. J. Mater. Chem. A 2016, 4, 2494−2500. (34) Xu, P.; Chen, S.; Xiang, H.-J.; Gong, X.-G.; Wei, S.-H. Influence of Defects and Synthesis Conditions on the Photovoltaic Performance of Perovskite Semiconductor CsSnI3. Chem. Mater. 2014, 26, 6068− 6072. (35) Tsai, C.-M.; Wu, H.-P.; Chang, S.-T.; Huang, C.-F.; Wang, C.H.; Narra, S.; Yang, Y.-W.; Wang, C.-L.; Hung, C.-H.; Diau, E. W.-G. The Role of Tin Chloride in Tin-Rich Mixed-Halide Perovskites Applied as Mesoscopic Solar Cells with Carbon Counter Electrode. ACS Energy Lett. 2016, 1, 1086−1093. (36) Yu, Y.; Zhao, D.; Grice, C.; Meng, W.; Wang, C.; Liao, W.; Cimaroli, A.; et al. Thermally Evaporated Methylammonium Tin Triiodide Thin Films for Lead-Free Perovskite Solar Cell Fabrication. RSC Adv. 2016, 6, 90248−90254. (37) Wang, F.; Ma, J.; Xie, F.; Li, L.; Chen, J.; Fan, J.; Zhao, N. Organic Cation-Dependent Degradation Mechanism of Organotin Halide Perovskites. Adv. Funct. Mater. 2016, 26, 3417−3423.
28
DOI: 10.1021/acsenergylett.6b00513 ACS Energy Lett. 2017, 2, 22−28
