Preferential (100)-oriented CH₃NH₃PbI₃ perovskite film formation by flash drying and elucidation of formation mechanism

Abstract

Most previous studies of perovskite films have explored the use of highly (110)-oriented perovskite films, even though films having the (100) orientation exhibit more desirable characteristics. In this study, we examined a simple method for growing (100)-oriented perovskite films for solar cells and elucidated their growth mechanism. Oriented perovskite grains with a lateral size of as much as 20 μm and a very flat surface morphology could be obtained. It was found that the amount of thermal energy delivered during annealing and the amount of residual solvent remaining after spin coating play critical roles in determining the growth orientation of the perovskite film. It was suggested that the formation mechanism of the preferentially (100)-oriented grains is most likely the classical hetero-nucleation and selective growth process in the solution system, even though DMSO was included in the solvent. The results of this study will aid in the optimization of preferentially (100)-oriented CH₃NH₃PbI₃ perovskite film, which will be useful for the study of the effect of crystal orientation and the properties of perovskite optoelectronic devices, such as LEDs, sensors, transistors, and photovoltaic cells.

---

Introduction

Recent interest in perovskite solar cells has led to attempts to optimize the morphology of perovskite films as well as the degree of perovskite crystallization, in order to not only maximize the transport parameters but also improve the charge-transfer properties and stability of the perovskite layer.^1–4 A number of strategies have been proposed for optimizing the perovskite layer. However, most of the related studies were performed on highly (110)-oriented perovskite films, even though (100)-orientation is also a good candidate orientation for application to perovskite solar cells.

The control of film morphology is very important because a smooth morphology with good surface coverage of the perovskite layer can reduce the shunting path and defect density, and can also improve the light absorption and charge transport properties.^1,5–9 The surface morphology of a perovskite film is determined by the orientation of its grains, because the exposed facets of the grains consist of planes with low surface energies. For example, the surface energies of the (100), (001), (110), (011), (101), (111), and (010) facets of orthorhombic perovskite have been reported as being 0.215, 0.226, 0.246, 0.246, 0.248, 0.252, and 0.279 J m⁻², respectively.⁹ Theoretically, orthorhombic perovskite films with the optimal morphology should be obtained when the growth is along the (100) plane.

It is also important to control the orientation of the perovskite film, in order to optimize not only the transport parameters of perovskite films but also their charge-transfer properties.³ For example, electron transfer to phenyl-C61-butyric acid methyl ester (PCBM), which is related to increases in the cell efficiency in the case of the inverted structure, can be optimized by using a (100)-oriented perovskite film.

Further, the orientation of perovskite films determines their degree of instability when brought into contact with water.⁴ It has been reported that water adsorption by perovskite is significantly affected by the orientation of the methyl ammonium (MA) cations close to the surface.⁴ In other words, the water resistivity could be improved by controlling the preferential orientation of the perovskite films. Hence, the control of the growth orientation of the perovskite films has great significance.

Interestingly, a recent study discussed the mechanism responsible for the orientation of perovskite films grown on anatase TiO₂.¹⁰ The study reported that the (110)-oriented preferential growth of perovskite was attributable to better structural matching between the perovskite and TiO₂.¹⁰ It is reasonable to assume that this has an effect on the morphology and hence the power conversion efficiency (PCE). This is because the orientation of the perovskite grains is related to the film morphology and the existence of strain on the perovskite grains due to the structural mismatch can cause changes in the electronic structure, resulting in changes in the effective carrier mass, exciton binding energy, and bandgap.¹¹ However, the growth mechanism of (110)-oriented perovskite films formed on TiO₂ layers is not yet known. Because the blocking TiO₂ layer consists of randomly oriented nanoparticles, the rearranged atoms on the surfaces of the TiO₂ nanoparticles are not located at the bulk lattice positions. The randomly oriented perovskite grains should be grown on the randomly oriented TiO₂ nanoparticles. Another clue regarding the preferential (110) orientation of perovskite films is that perovskite films primarily exhibiting the (110) orientation are formed regardless of the coating method or substrate used.^12–22 For example, the (110)-oriented preferential growth of perovskite is observed in the case of a number of growth methods, such as the solvent engineering method,^12–15 the solvent–solvent extraction method,¹⁶ the sequential-dipping method,^17–19 and the vapor-deposition method.²² These two facts suggest that the (110) orientation of perovskite is not dependent on the interaction between the substrate and the formed film.

In this article, we describe a simple method for growing (100)-oriented perovskite thin films and elucidate their growth mechanism. The method involves a one-step spin coating process and subsequent annealing at a high temperature for a few seconds. It was observed that the orientation and coverage of the thus-formed films changed dramatically with the spinning time and annealing temperature. Also, it can be suggested that the formation mechanism of the preferentially (100)-oriented grains is most likely the classical hetero-nucleation and selective growth process in the solution system, even though DMSO was included in the solvent.

Results and discussion

Fig. 1 shows the log-scale X-ray diffraction (XRD) patterns of the perovskite films formed on blocking-TiO₂ (b-TiO₂)/fluorine-doped tin oxide (FTO) substrates, as determined from powder-mode measurements. The XRD patterns of the b-TiO₂ layer on the FTO substrate and the PbI₂ film are also shown, to enable a comparison of the peak origins of the perovskite phase; the peak of the PbI₂ film was not observed in the XRD patterns of the perovskite films. Fig. 1(a) shows the XRD patterns of perovskite films that were spin-coated for 5 s and annealed at 100, 150, 200, 300, and 400 °C for 5 min, 2 min, 5 s, 3 s, and 3 s, respectively. Fig. 1(b) shows the XRD patterns of the perovskite films formed by spinning for 5, 30, and 60 s and annealed at 400 °C. It was assumed that the temperature of the sample would be substantially lower than the set temperature during the annealing process, because the short annealing time was not long enough to reach the set temperature. In addition, the substrate thickness was 2.3 mm; we wanted to describe the thermal energy in terms other than the annealing temperature. However, it proved difficult to measure the specific thermal energy which was applied to the wet films on the thick substrate for few seconds. Therefore, the hotplate temperature was used as the annealing temperature in this study.

Fig. 1 XRD patterns of perovskite films formed on a b-TiO₂ layer at (a) annealing temperatures of 100, 150, 200, 300, and 400 °C and for (b) spinning times of 5, 30, and 60 s at 400 °C.

In the case of the film annealed at 100 °C, a dominant (110) perovskite peak was observed, as shown in Fig. 1(a). The intensities of the (200) and (400) peaks dramatically increased in the case of those films annealed at 150, 200, 300, and 400 °C, as shown in Fig. 1(a). It can be concluded that the amount of thermal energy supplied to the wet films should be sufficient to grow (100)-oriented perovskite films. Further, a dominant (200) peak and (400) peak related to perovskite were observed in the case of those films formed with spinning times of 5 and 30 s and an annealing temperature of 400 °C, while the (110) peak was dominant in the case of the films formed with a spinning time of 60 s, as shown in Fig. 1(b). This suggested that a short spinning time, provided sufficient thermal energy is applied to the wet film, should be applied to grow (100)-oriented perovskite films. In other words, the amount of residual solvent plays a significant role in determining the orientation of the perovskite films.

Interestingly, (100)-oriented perovskite films could be obtained readily on a different substrate, namely, a poly(3,4-ethylenedioxythiophene) polystyrene sulfonate (PEDOT:PSS)-coated indium tin oxide (ITO) substrate, as shown in Fig. S1 and S2,† using a short spinning time and a relatively high annealing temperature. This suggested that the film orientation was not closely related to the type of underlying layer used.

Fig. 2 shows field emission scanning electron microscopy (FESEM) images of perovskite films annealed at 100, 150, 200, 300, and 400 °C for 5 min, 2 min, 5 s, 3 s, and 3 s, respectively. The microstructures of the perovskite films could be divided into three types, based on their preferential orientation and degree of coverage, as shown in Fig. 2. The first type (see Fig. 2(a)–(d)) corresponded to a rough morphology, with the perovskite films exhibiting uncovered areas. This microstructure was observed in the films annealed at 100 °C for 5 min. A b-TiO₂ layer can be observed in the top view of the perovskite films. Further, the perovskite grains were randomly agglomerated on the b-TiO₂ layer.

Fig. 2 FESEM images, with different magnifications, of perovskite films formed by annealing at (a)–(d) 100 °C; (e)–(h) 150 °C; (i)–(l) 200 °C; (m)–(p) 300 °C; and (q)–(t) 400 °C.

The second type of microstructure, which was a flower-like microstructure with an uncovered area, was observed in the FESEM images of the films annealed at 150 °C, 200 °C, and 300 °C, as shown in Fig. 2(e)–(h), (i)–(l) and (m)–(p), respectively. Interestingly, the grains of the flower-like microstructure were of two types: (i) small grains, which were observed in the center of the flower-like microstructure, and (ii) large grains, which were located on the outside of the flower-like microstructure, as shown in Fig. 3(b). (2) The b-TiO₂ layer could be seen in the top-view of the FESEM images of the films, as was also the case for the films with the first type of microstructure (see Fig. 2(g), (k) and (o)).

Fig. 3 (a) Low-magnification FESEM images of perovskite film formed by 5 s spinning time and annealing at 200 °C for 5 s, (b) overlapping FESEM image of white area in (a). Cross-sectional views of (c) outer and (d) inner grains (another overlapped FESEM image is shown in Fig. S3†).

In the case of the third type of microstructure, a flower-like microstructure without an uncovered layer was observed, as shown in Fig. 2(q)–(t). The third type of microstructure was similar to the second type; they differed in that the b-TiO₂ layer was not observed in the former case.

Based on these results, it was concluded that the evaporation rate of the solvents is a key factor for controlling the growth and orientation of perovskite films during the annealing process. This is because the annealing temperature is closely related to the rate of solvent evaporation. This was particularly true for those perovskite films with the second and third types of microstructures; for these films, the annealing time was in the order of a few seconds. If a solvent having a low evaporation rate (low vapor pressure) is used, the wet film state can be maintained for a longer duration at a high annealing temperature. According to the extended duration of the wet-film state at high annealing temperatures, the highly (100)-oriented perovskite grains could be grown easily. A schematic of the growth mechanism of perovskite films grown using dimethyl sulfoxide (DMSO) and gamma butyrolactone (GBL) as the solvents is shown in Fig. 5.

Fig. 4 FESEM images of perovskite films formed by annealing at 400 °C for spinning times of (a)–(c) 5 s; (d)–(f) 30 s; and (g)–(i) 60 s.

Fig. 5 Schematics of the (a) three types of perovskite film microstructures formed at the different annealing temperatures and spinning times. Microstructures of the (b) randomly oriented film with an uncovered area, (c) preferentially (100)-oriented film with an uncovered area, and (d) preferentially (100)-oriented film without an uncovered area; these correspond to the first, second, and third types of microstructures, respectively.

Fig. 3 shows an overlapping FESEM image of the perovskite films annealed at 200 °C. The films annealed at 150, 200, and 300 °C all exhibited similar microstructures, that is, the second type of microstructure. Fig. 3(b) shows a high-magnification image of the white area in Fig. 3(a). This shows another overlapping image, extending from the inside of the flower-like microstructure to the outside; the outside is characterized by highly (100)-oriented large grains outside the flower-like microstructure, while the inside consists of randomly oriented small grains at the center of the flower-like microstructure, resembling actual chamomile ray and disk flowers, respectively. These are shown in Fig. 5(c). Further, the uncovered area of the film is observed at the end of the large grains located outside, as shown in Fig. 3(b). The largest lateral grain size was in excess of 20 μm, with the grains exhibiting very flat surfaces. It was thought that the large grains might be due to the selective grain growth during the flash annealing process. The mechanism whereby abnormal grain growth occurs is described in Fig. 5. Cross-sectional views of the outer and inner grains are shown in Fig. 3(c) and (d), respectively; it can be seen that the outer grains are flatter than the inner ones. The larger grains are probably (100)-oriented grains, because the b-TiO₂ layer was covered primarily with the larger grains and the main peaks in the XRD patterns were related to the (200) and (400) planes of perovskite. In general, the preferential growth of a crystal structure is along the plane having the highest surface energy. Thus, it was expected that the direction for the lateral grain growth of the perovskite films would be along the (0kl) plane, which has the highest surface energy in the case of tetragonal perovskite.

Fig. 4 shows FESEM images of the perovskite films formed with spinning times of 5, 30, and 60 s and an annealing temperature of 400 °C. A preferentially (100)-oriented film without any uncovered areas and having the third type of microstructure was formed when the spinning time was 5 s, as shown in Fig. 4(a)–(c). On the other hand, a preferentially (100)-oriented film with an uncovered area and having the second type of microstructure was formed when the spinning time was 30 s, as shown in Fig. 4(d)–(f). Finally, a randomly oriented film with an uncovered area and having the first type of microstructure was observed when the spinning time was 60 s, as shown in Fig. 4(g)–(i). Based on these results, it was concluded that the amount of wet-state residual solvent plays a very important role in determining the orientation and microstructure of perovskite films.

As mentioned above, the microstructures of the perovskite films could be divided into three types, based on the annealing temperature and spinning time, as shown in Fig. 5(a). In the case of the first type, randomly oriented films with uncovered areas were formed for a low annealing temperature and a long spinning time. In the case of the second type, preferentially (100)-oriented films with an uncovered area were formed at a moderately high annealing temperature and medium spinning time. In the case of the third type, preferentially (100)-oriented films without any uncovered areas were formed at a high annealing temperature and short spinning times. Fig. 5(b)–(d) show illustrations of the growth behaviors of the three types of microstructures.

The growth behavior of the randomly oriented film with an uncovered area and having the first type of microstructure is shown in Fig. 5(b). It is known that the final microstructure of perovskite films is dependent on the microstructure of the intermediate methylammonium iodide (MAI)–PbI₂–DMSO phase.¹² The mechanism whereby the microstructure of the intermediate MAI–PbI₂–DMSO phase is formed could be explained by the intermolecular force between the solvents used and the state of the complex during the drying process. The DMSO molecule, which forms an intermolecular bond with the PbI₆ octahedral layer, exhibits a large dipole moment (3.96 D). Further, clusters of MAI–PbI₂–DMSO would also have a dipole moment because the MAI–PbI₂–DMSO complex consists of oriented DMSO molecules. Hence, the intermolecular force attributable to the dipole moments of the solvent and the clusters leads to the formation of specific microstructures through the anisotropic adhering of the intermediate phases, such as needle-shaped microstructures. For example, films of the intermediate MAI–PbI₂–DMSO phase have been frozen using nonpolar solvents such as toluene, chlorobenzene, and diethyl ether through a method known as solvent engineering, in order to form dense perovskite films.^12–15,22 However, perovskite films that have not been frozen (such as those shown in Fig. 2(a) and 4(g)) commonly exhibit a rough morphology, since only polar aprotic solvents are used in this case.¹⁴ Based on these results, it is possible to conclude that the randomly oriented films with uncovered areas originated from a semi-dried MAI–PbI₂–DMSO complex's microstructure and that the microstructures of the semi-dried complex phase were formed during the solvent-drying process, which was performed under conditions that included a low annealing temperature and/or a long spinning time. Since the rotation and diffusion of the MAI–PbI₂–DMSO clusters for thermodynamically stable adhering can occur in the wet-film state, randomly oriented complex films with needle-shaped microstructure can be formed, given the characteristics of DMSO as a solvent, such as its low vapor pressure, high boiling point, and high dipole moment; the vapor pressure of DMSO at 20 °C is 0.42 Torr and its boiling point is 189 °C.

In the case of the second type of microstructure, the growth behavior of the preferentially (100)-oriented films with uncovered areas is as shown in Fig. 5(c). The preferentially (100)-oriented perovskite phase could not be formed through the MAI–PbI₂–DMSO intermediate phase. We propose that the preferentially (100)-oriented grains are formed through classical heteronucleation and selective growth in the solution system.

It was assumed that randomly oriented seeds of perovskite are initially formed on the b-TiO₂ layer by heteronucleation, as shown in S4,† even though a DMSO solvent which produces a stable complex phase was used. It was expected that heteronucleation occurs regardless of the existence of DMSO solvents when sufficient thermal energy was induced in a wet film; additional investigation should be done to distinguish the origin of the heteronucleation in a liquid/solid system, because the DMSO solvents have inverse solubility properties.²³ After the heteronucleation, it was suggested that the abnormal growth of the (100)-oriented perovskite grains occurs by selective growth at the liquid/solid interface; the growth of grains having a specific orientation is usually achieved by selective growth on the different planes having different surface energies. The proposed growth model assumed that the dissolved perovskite components in the solvent diffused quickly and freely to the plane having the highest surface energy during the short annealing process; the existence of liquid was important for growth. An additional nucleation process would be required for growth on a plane having a low surface energy, such as a plane without a terrace, ledge, or kink sit. However, growth can occur readily even in the absence of an additional nucleation process on the plane having a high surface energy, such as a plane with many terraces, ledges, and kink sites, owing to its unstable surface. In other words, among the various randomly oriented nuclei, the (100)-oriented grains would grow readily along the 〈0kl〉 direction.

In the case of the third type of microstructure, the growth behavior of the preferentially (100)-oriented films without uncovered areas is as shown in Fig. 5(d). The formation mechanism of the films with the third type of microstructure was similar to that of the films with the second type of microstructure. The difference between these was that fully covered films were realized at higher annealing temperatures. Thus, it was concluded that preferentially (100)-oriented films without any uncovered area were formed in the highly supersaturated environment resulting from the use of a high annealing temperature. It could be expected that there would be competition between the lateral grain growth of the (100)-oriented grains and additional nucleation on the surface of the substrate owing to supersaturation at 400 °C, as explained in S5.† Interestingly, based on these results, it can be expected that annealing temperatures greater than 400 °C will result in randomly oriented films without any uncovered areas, as illustrated in Fig. S6 and S7.†

Our suggested formation mechanism can be understood in the same context as that used to describe grain growth by the hot casting method using a halogen lamp; a halogen lamp was used to apply thermal energy for a few seconds.²⁴ It was found that grain growth was dependent on the annealing temperature (130–180 °C) and the boiling point of the solvents. In other words, the thermal energy and remaining amount of solvent played important roles in grain growth. However, the results did not exhibit any preferentially oriented film. Rather, only randomly oriented film with a fully covered area was reported. Interestingly, it was expected that the thermal energy of our experimental conditions (400 °C) would be lower than under the above-mentioned conditions (130–180 °C), because of the shorter annealing time (3 s), thicker FTO substrate (2.3 mm), low temperature (room temperature) of the wet film, and the latent heat by evaporation of the solvent mixture (DMSO and GBL). Hence, we could assume that this is the reason why the PbI₂ phase was not observed in our samples; it is well known that the MAPbI₃ does not have good thermal stability, making the heat-treatment of a MAPbI₃ wet film at high temperature for an extended period impractical.²⁵ In other words, the real thermal energy was important for growing (100)-oriented films; for example, when using a thin (0.5 mm) ITO substrate, the (100)-oriented perovskite film was easily formed at an annealing temperature of 100 °C, as shown in Fig. S1 and S2.†

Table 1 lists the average device parameters while Fig. S8† shows the characteristic photocurrent–voltage plots of perovskite solar cells based on the films annealed at different temperatures (100, 200, 300, and 400 °C). Under standard global AM1.5 solar irradiation, the device with the film annealed at 400 °C showed a short-circuit current density (J_SC) of 17.82 mA cm⁻², an open-circuit voltage (V_OC) of 0.89 V, and a fill factor (FF) of 48.27%; these correspond to an overall conversion efficiency (η) of 7.62%. Under the same conditions, other devices exhibited lower J_SC, FF, and V_OC values. Thus, the efficiency of the device with the film annealed at 400 °C was the highest. This can be explained on the basis of the fact that the extent of coverage of the corresponding perovskite film was the highest. Thus, it can be concluded the low PCE value of the thus-fabricated devices can be improved by optimizing the device fabrication process as well as the thickness of the light-absorbing film.

Table 1 Parameters of perovskite solar cells based on films formed by annealing at 100, 200, 300, and 400 °C and spinning for 5 min, 5 s, 3 s, and 3 s, respectively

Temp.	VOC (V)	JSC (mA cm^−2)	FF (%)	η (%)
100 °C	0.70	10.86	20.43	1.56
200 °C	0.81	17.6	35.25	5.26
300 °C	0.83	16.72	36.9	5.24
400 °C	0.89	17.82	48.27	7.62

---

Experimental

MAI synthesis

CH₃NH₃I was synthesized by the reaction of 27.8 ml of CH₃NH₂ (40 wt% in methanol, TCI) with 30 ml of hydrogen iodide (HI, 57 wt% in water, Aldrich) in a round flask while stirring for 2 h in an ice bath. The resulting compound, CH₃NH₃I, was collected using a rotary evaporator at 50 °C for 1 h, washed with diethyl ether several times, and then dried in a vacuum for 24 h.

Perovskite film and solar device fabrication

FTO substrates (Pilkington, TEC-8, 8 Ω sq⁻¹) were cleaned with acetone, ethanol, and isopropyl alcohol in an ultrasound cleaner for 15 min each and then treated with an ultraviolet ozone cleaner for 15 min. A 30 nm-thick b-TiO₂ layer was then formed by spin coating a 0.15 M solution of titanium diisopropoxide bis(acetylacetonate) (Sigma-Aldrich, 75 wt% in isopropanol) in 1-butanol (Sigma-Aldrich, 99.8%) at 3000 rpm for 30 s. The resulting films were annealed at 450 °C for 1 h.

Next, 1 M solutions of PbI₂ and MAI were prepared in DMSO and GBL (DMSO/GBL = 3 : 7) at room temperature for 12 h. Next, two experiments were performed to investigate the growth orientation of the perovskite films, as follows.

The effect of the annealing temperature on the growth orientation of the perovskite was as follows: a 1 M solution of PbI₂ and MAI was spin-coated on the b-TiO₂ layer at 3000 rpm for 5 s. After the completion of the spin coating process, the wet films were placed on a hotplate and heated to 100, 150, 200, 300, and 400 °C for 5 min, 2 min, 5 s, 3 s, and 3 s, respectively. The various annealing times at different annealing temperatures were applied to minimize the decomposition from perovskite to the PbI₂ phase under each condition. For temperatures higher than 200 °C, flash annealing was performed in order to prevent the perovskite coating from decomposing into PbI₂. For example, a perovskite film without traceable PbI₂ was formed after heating a wet film to 400 °C for 3 s. However, PbI₂ was observed because of phase decomposition after annealing for 5 s.

It was assumed that the temperature of the sample would be substantially lower than the set temperature during the annealing process because of the short annealing time, thickness of the FTO substrate (2.3 mm), low temperature (room temperature) of the wet film, and latent heat by evaporation of the mixed solvents (DMSO and GBL).

The effect of the residual wet film solvents on the growth orientation of the perovskite films was as follows: the 1 M solution of PbI₂ and MAI was spin-coated on the b-TiO₂ layer at 3000 rpm for 5, 30, and 60 s. After the spinning process, the films were annealed at 400 °C for 3 s.

The hole-transport material (HTM) was deposited on the substrate as follows: 200 μl of a spiro-MeOTAD solution was spin-coated on the CH₃NH₃PbI₃ layer at 4000 rpm for 20 s. The spiro-MeOTAD solution was prepared by dissolving 72.3 mg of spiro-MeOTAD in 1 ml of chlorobenzene, 28.8 μl of 4-tert-butyl pyridine, and 17.5 μl of a lithium bis(trifluoromethanesulfonyl) imide (Li-TFSI) solution (520 mg of Li-TFSI in 1 ml of acetonitrile, Sigma-Aldrich, 99.8%). Finally, a 50 nm Au counter electrode was deposited by thermal evaporation on the top of the HTM layer under an approximately 10⁻⁶ Torr vacuum at a deposition rate of approximately 0.1 nm s⁻¹.

Film and device characterization

The photocurrent and voltage were measured with a solar simulator equipped with a 450 W xenon lamp (Newport 91192-1000) and a Keithley 2400 source meter. The light intensity was adjusted with an NREL-calibrated Si solar cell with an intensity of approximating 1 sun. While measuring the current and voltage, the cell was covered with a black mask with an aperture (the aperture area was 0.096 cm²).

The morphologies and microstructures of the perovskite films were characterized using (FE-SEM). The crystal structures of the perovskite films were investigated by using a high-resolution XRD (HR-XRD) measurement system with a Cu Kα radiation source operated at 40 kV and 30 mA.

Conclusions

In this article, we have described a simple method for growing (100)-oriented perovskite films and suggest their growth mechanism. It was found that the amount of thermal energy delivered to the films and the amount of solvent remaining after the spin-coating process play key roles in controlling the growth of (100)-oriented perovskite films. Additionally, the substrate treatment may play an important role in the optimization of grain growth; it was reported that the grain growth can be improved by using the non-wetting surface of a substrate due to the weak bonding between the film and substrate.²⁶

Based on our results, we propose that the formation mechanism of the preferentially (100)-oriented grains is related to classical heteronucleation and selective growth in the solution system, even though DMSO was used as one of the solvents. In other words, we suggest that the nucleation and growth occurs directly from liquid to solid, not through the intermediate phase, such as MAI–PbI₂–DMSO. As a result, fully covered highly (100)-oriented perovskite films could be formed using a short spinning time and high annealing temperature without the need for cumbersome processes, such as the solvent engineering method.

However, solar cells based on the fabricated (100)-oriented perovskite films have not been optimized because this study focused on the formation of (100)-oriented MAPbI₃ perovskite film. In the future, we intend to optimize the growth of the (100)-oriented perovskite films by using different solvents (different vapor pressure and latent heat) and by controlling the ratio at which the solvents are mixed. Subsequently, the stability of the resulting (100)-oriented perovskite films will be characterized.

Acknowledgements

This work was supported by the DGIST R&D Program of the Ministry of Education, Science and Technology of Korea (16-EN-03) and Development of Semi-transparent Wide Bandgap Solar Cells (16-HRMA-02).

References

1. G. E. Eperon, V. M. Burlakov, P. Docamp, A. Goriely and H. Snaith, Adv. Funct. Mater., 2014, 24, 151.
2. P. W. Liang, C. Y. Liao, C. C. Chueh, F. Zuo, S. T. Williams, X. K. Xin, J. Lin and A. K. Y. Jen, Adv. Mater., 2014, 26, 3748.
3. J. Yin, D. Cortecchia, A. Krishna, S. Chen, N. Mathews, A. C. Grimsdale and C. Soci, J. Phys. Chem. Lett., 2015, 6, 1396.
4. N. Z. Koocher, D. Saldana-Greco, F. Wang, S. Liu and A. M. Rappe, J. Phys. Chem. Lett., 2015, 6, 4371.
5. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y.-B. Cheng and L. Spiccia, Angew. Chem., Int. Ed., 2014, 53, 9898.
6. H. J. Snaith, A. Abate, J. M. Ball, G. E. Eperon, T. Leijtens, N. K. Noel, S. D. Stranks, J. T.-W. Wang, K. Wojciechowski and W. Zhang, J. Phys. Chem. Lett., 2014, 5, 1511.
7. H.-S. Kim and N.-G. Park, J. Phys. Chem. Lett., 2014, 5, 2927.
8. Y. Shao, Z. Xiao, C. Bi, Y. Yuan and J. Huang, Nat. Commun., 2014, 5, 5784.
9. Y. Wang, B. G. Sumpter, J. Huang, H. Zhang, P. Liu, H. Yang and H. Zhao, J. Phys. Chem. C, 2015, 119, 1136.
10. E. Mosconi, E. Ronca and F. D. Angelis, J. Phys. Chem. Lett., 2014, 5, 2619.
11. M. R. Filip, G. E. Eperon, G. J. Snaith and F. Giustino, Nat. Commun., 2014, 5, 5757.
12. N. J. Jeon, J. H. Noh, Y. C. Kim, W. S. Yang, S. C. Ryu and S. I. Seok, Nat. Mater., 2014, 13, 897.
13. J. W. Jung, S. T. Williams and A. K. Y. Jen, RSC Adv., 2014, 4, 62971.
14. M. Xiao, F. Huang, W. Huang, Y. Dkhissi, Y. Zhu, J. Etheridge, A. Gray-Weale, U. Bach, Y. B. Cheng and L. Spiccia, Angew. Chem., 2014, 126, 10056.
15. Y. J. Jeon, S. Lee, R. Kang, J. E. Kim, J. S. Yeo, S. H. Lee, S. S. Kim, K. M. Yun and D. Y. Kim, Sci. Rep., 2014, 4, 6953.
16. Y. Zhou, M. Yang, W. Wu, A. L. Vasiliev, K. Zhu and N. P. Padture, J. Mater. Chem. A, 2015, 3, 8178.
17. T. Zhang, M. Yang, Y. Zhao and K. Zhu, Nano Lett., 2015, 15, 3959.
18. H. S. Ko, J. W. Lee and N. G. Park, J. Mater. Chem. A, 2015, 3, 8808.
19. H. Zheng, W. Wang, S. Yang, Y. Liu and J. Sun, RSC Adv., 2016, 6, 1611.
20. J. H. Heo, D. H. Song, H. J. Han, S. Y. Kim, J. H. Kim, D. Kim, H. W. Shin, T. K. Ahn, C. Wolf, T. W. Lee and S. H. Im, Adv. Mater., 2015, 27, 3424.
21. J. H. Heo, H. J. Han, D. Kim, T. K. Ahn and S. H. Im, Energy Environ. Sci., 2015, 8, 1602.
22. N. Ahn, D. Y. Son, I. H. Jang, S. M. Kang, M. Choi and N. G. Park, J. Am. Chem. Soc., 2015, 137, 8696.
23. M. I. Saidminov, A. L. Abdelhady, B. Murali, E. Alarousu, V. M. Burlkov, W. Peng, I. Dursun, L. Wang, Y. He, G. Maculan, A. Goriely, T. Wu, O. F. Mohammend and O. M. Bakr, Nat. Commun., 2015, 6, 7586.
24. W. Nie, J. Tsai, R. Asadpour, J.-C. Blancon, A. J. Neukirch, G. Gupta, J. J. Crochet, M. Chhowalla, S. Tretiak, M. A. Alam, H.-L. Wang and A. D. Mohite, Science, 2015, 347, 522.
25. Z. Song, S. C. Watthage, A. B. Phillips, B. L. Tompkins, R. J. Ellingson and M. J. Heben, Chem. Mater., 2015, 27, 4612.
26. C. Bi, Q. Wang, Y. Shao, Y. Yuan, Z. Xiao and J. Huang, Nat. Commun., 2015, 6, 7747.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra21423a

---