Surface Properties of CH3NH3PbI3 for Perovskite Solar Cells

Feb 22, 2016 - †Global Research Center for Environment and Energy Nanoscience (GREEN), and ‡International Center for Materials Nanoarchitectonics,...
0 downloads 10 Views 2MB Size
Article pubs.acs.org/accounts

Surface Properties of CH3NH3PbI3 for Perovskite Solar Cells Published as part of the Accounts of Chemical Research special issue “Lead Halide Perovskites for Solar Energy Conversion”. Jun Haruyama,*,† Keitaro Sodeyama,‡,§ Liyuan Han,∥,⊥ and Yoshitaka Tateyama*,†,‡,§,⊥ †

Global Research Center for Environment and Energy Nanoscience (GREEN), and ‡International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305- 0044, Japan § Elements Strategy Initiative for Catalysts and Batteries, Kyoto University, Goryo-Ohara, Nishikyo-ku, Kyoto 615-8245, Japan ∥ Photovoltaic Materials Unit, National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan ⊥ PRESTO and CREST, Japan Science and Technology Agency, 4-1-8 Honcho, Kawaguchi, Saitama 333-0012, Japan CONSPECTUS: Perovskite solar cells (PSCs) have attracted considerable interest because of their high potential for solar energy conversion. Power conversion efficiencies of the PSCs have rapidly increased from 3.8 to over 20% only in the past few years. PSCs have several similarities to dye-sensitized solar cells in their device compositions; mesoporous TiO2 (mp-TiO2) is sensitized by light-absorbing components and placed into a medium containing hole transporting materials (HTMs). On the other hand, the perovskite materials for the light-harvesting, for example, CH3NH3PbI3 (MAPbI3), have a greater advantage for the photovoltaic applications; extremely long photocarrier diffusion lengths (over 1 μm) enable carrier transports without singnificant loss. In this respect, the surface states, that can be possible recombination centers, are also of great importance. Availability of solution processes is another important aspect in terms of low cost fabrication of PSCs. Two-step methods, where PbI2 is first introduced from solution onto a mp-TiO2 film and subsequently transformed into the MAPbI3 by the exposition of a solution containing MAI, suggest that use of such a high PbI2 concentration is crucial to obtain higher performance. The experiments also indicate that the PbI2-rich growth condition modifies TiO2/ or HTM/MAPbI3 interfaces in such a way that the photocarrier transport is improved. Thus, the characteristics of surfaces and interfaces play key roles in the high efficiencies of the PSCs. In this Account, we focus on the structural stability and electronic states of the representative (110), (001), (100), and (101) surfaces of tetragonal MAPbI3, which can be regarded as reasonable model HTM/MAPbI3 interfaces, by use of first-principles calculations. By examining various types of PbIx polyhedron terminations, we found that there are two major phases on all of the four surface facets. They can be classified as vacant- and flat-type terminations, and the former is more stable than the latter under thermodynamically equilibrium conditions. More interestingly, both terminations can coexist especially on the more probable (110) and (001) surfaces. Electronic states, that is, projected density of states, of the stable-vacant and PbI2-rich-flat terminations on the two surfaces are almost the same as that in bulk MAPbI3. These surfaces can contribute to the long carrier lifetime actually observed for the PSCs because they have no midgap surface states. Furthermore, the shallow surface states on the (110) and (001) flat terminations can be efficient intermediates for hole transport to HTMs. Consequently, the formation of the flat terminations under the PbI2-rich condition will be beneficial for the improvement of PSC performance.



INTRODUCTION

solid hole transporting materials (HTMs), an instability of deposited MAPbI3 in liquid electrolyte was improved. Furthermore, the perovskite solar cells (PSCs) with longterm stability recorded an increased efficiency of 9.7%.3 After just 5−6 years, PCEs of PSCs have increased from the 3.8 to over 20%.1−16 This rapid increase can be interpreted as distinctive characters of organolead halide perovskites for photovoltaic applications compared with conventional DSSCs.17−20 A

Energy conversion of sunlight into electricity is absolutely essential for renewable energy and sustainable economic growth without negative environmental impact. Recently, solar cells composed of organometal halide perovskites have been intensively investigated because of rapid growth of their power conversion efficiencies (PCEs). Miyasaka et al. first attempted CH3NH3PbX3 (MAPbX3, X = Br, I) as a light harvester in liquid-based dye sensitized solar cells (DSSCs) in 2009, where the iodide cell showed PCE of 3.8%.1 Higher efficiency (6.5%) was achieved by Park et al. employing MAPbI3 and iodide redox electrolyte in 2011.2 Combined with © XXXX American Chemical Society

Received: October 4, 2015

A

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research primary advantage is that they strongly absorb sunlight over a broader range enabling the thinner sensitized layer.21,22 An optical absorption coefficient of MAPbI3 is greater than 105 cm−1 in short wavelength region.23 More interestingly, the organolead halide perovskites show long diffusion lengths approaching or exceeding 1 μm for both charge carriers.24−27 High mobility and slow recombination rate of holes and electrons enable the remarkable intrinsic properties in the organolead halide perovskites.28−30 It should be mentioned that photoexcited electron−hole pairs acts like free carriers lather than generally accepted exciton model. Femtosecond transient absorption spectroscopy measurements indicate that the dominant relaxation occurs through recombination of free electrons and holes.31,32 The remarkable optoelectronic characters are theoretically elucidated by their electronic structures,33 including a suitable band gap, strong transition dipole moment,34 low effective masses of the carriers,35−37 and shallow defect levels38−40 even at surfaces40,41 or grain boundaries.34 Especially, the absence of additional midgap states leads to a few carrier traps and nonradiative recombination centers, and preserves carrier delocalization for a certain amount, which results in the long diffusion length. In this respect, energy levels of the surface states are of great importance as well, and the main concern of this Account. From perspective of sufficiently low cost of fabrication, solution processes are highly desirable. Because PSCs are derived from DSSCs, the two cells have similarities in device compositions; light-absorption components are placed between carrier-transfer mediums such as a mesoporous-TiO2 (mpTiO2). A one step method, for example, spin coating for introducing perovskite components onto a carrier-transfer layer, was applied at first.1,42 Later, high performance cells are effectively provided by a two-step method; PbI2 is deposited from solution onto a mp-TiO2 film and then transformed into the MAPbI 3 . 5,13 We suppose that such a high PbI 2 concentration, that is, the PbI2-rich growth condition, is favorable for the MAPbI3 surface, TiO2/MAPbI3, and HTM/ MAPbI3 interfaces in such kind of situation as to improve photoexcited carrier transport. In this Account, we focus on electronic properties, especially surface states, of various terminations of the tetragonal MAPbI3 surfaces from a viewpoint of growth conditions, which are substituted for different chemical potentials, by using firstprinciples calculations. Note that most of the results have been reported in our earlier publication.41 The provided surfaces are reasonable interface models of MAPbI3/HTMs represented as spiro-OMeTAD (2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-bifluorene) owing to the weak interactions between them. Desired properties for good PSC performances will be suggested with consequent stable terminations and surface electronic states. Finally, future prospects are presented for theoretical surface and interface investigations based on the experimental findings.

Figure 1. (a) Correspondence between the tetragonal and cubic (conventional ABX3) cells. Red circles represent A site, while B and X sites place vertexes of the black and purple square, respectively. (b) Tetragonal phase of MAPbI3. The H, C, N, I, and Pb atoms are expressed as white, brown, light blue, purple, and black spheres, respectively. PbI6 units are depicted by black octahedron. Blue dashed lines indicate the representative (110), (100), (101), and (001) planes. Adapted with permission from ref 41. Copyright 2014 American Chemical Society.

conventional unit cell (Figure 1b, crystal structures are described by VESTA46). In this Account, density functional theory (DFT) calculations were conducted on the simple (110), (001), (100), and (101) surfaces of tetragonal MAPbI3. Figure 1a shows relationship between the cubic and tetragonal structures. Directions of tetragonal [110], [001], [100], and [101] are identical to these of cubic ABX3 [010], [001], [110], and [111], respectively. As illustrated Figure 1b, the tetragonal (110) and (001) planes constructed with the neutral [MAI]0/ [PbI2]0 layers, namely they are nonpolar surfaces. On the other hand, the tetragonal (100) and (101) surfaces are consist of the charged [MAPbI]2+/[I2]2− and [MAI3]2−/[Pb]2+ layers, which surfaces may need reconstructions or defect formations. Therefore, (110) and (001) surfaces are anticipated to be more stable than (100) and (101). Experimentally, tetragonal MAPbI3 on mp-TiO2 films show X-ray diffraction peaks of the (110) and (001),47 while planar heterojunction PSCs consisting of MAPbI3−xClx show the (110) peaks only.6 Note that a recent experiment indicates Cl depletion at surface region,48 whereas the role of Cl is still under debate.49 After investigation of structural stability, therefore, we introduced electronic properties of the probable (110) and (001) surfaces. A number of terminations on the four target surfaces, where they have two or four PbI6 octahedrons in the primitive cell due to the tetragonal phase, was investigated as well as flat terminations. Components of the outermost PbIx polyhedron layers are used to classify each termination. As depicted in Figure 2, a flat termination of the (110) surface is labeled as -(PbI5)4. The flat plane with additional 4 I atoms is labeled as -(PbI6)4, and another flat plane with missing 2 (PbI2) unit is -(PbI5)(PbI3). Eventually, five to six terminations have been investigated on the four individual surfaces (all of the terminations investigated are listed in ref 41). To avoid failures of DFT calculations emerged with net dipole moment in the supercell, the same terminations on both sides of the MAPbI3 slab were treated.



PEROVSKITE SURFACES AND COMPUTATIONAL MODELS Perovskite has a unit cell represented as ABX3 formula in a cubic structure (Figure 1a shows the top view of the cubic structure), here A, B, and X represent cations (e.g., MA, Pb) and anions (e.g., Br, I), respectively. In the case of MAPbI3, the most representative material for PSC, tetragonal phase with lower symmetry is stable at room temperature.43,44 The tetragonal phase45 consists of a √2 × √2 × 2 ABX3 B

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research ΔHform[MAPbI3] ≤ ΔμPb ≤ 0, ΔHform[MAPbI3] ≤ ΔμI ≤ 0

(3)

ΔHform[MAPbI3] − ΔHform[MAI] ≤ ΔμPb + 2ΔμI ≤ ΔHform[PbI 2]

(4)

where ΔHform is a heat of formation. They are represented as the difference between total energy of a compound and composition-weighted sum of their constituents, ΔHform[AB] = Etot[AB] − Etot[A] − Etot[B]

Figure 2. Schematic picture for the construction of various terminations. The outside rectangles represent boundaries of a periodic DFT cell including vacuum layer.

(5)

The values of ΔHform (MAPbI3, MAI, PbI2)= −5.49, −3.02, and −2.39 eV are derived from our PBE 57 based DFT calculations.41 Range of chemical potentials defined by eqs 3 and 4 is shown in Figure 3, which is almost the same as the

A previous experiment observed a small understoichiometry of N and I atoms synthesized by the two-step method (reported composition ratio is N:I = 1:2.8);50 therefore, we did not consider the MA terminations. In atomic relaxation procedure, initial MA configurations were decided to reference of a previous study by Mosconi et al.51 In addition, to avoid unreasonable large dipole in calculated supercell, we canceled out the surface polarization to exchange the C and N atoms in a symmetric way. It is mentioned that the rotated and migrated motion (they could be related to the ferroelectricity and ion conduction, respectively) of A site molecules in metal halide perovskites are currently main concerns about hysteresis and stability of PSCs. Thus, further computational studies are necessary for understanding of these behaviors.52−54



PHASE DIAGRAMS AND STABLE SURFACE STRUCTURES To investigate structural stabilities of a variety of different stoichiometries, PbαIβ(MAPbI3)γ, we applied the grand potential Ω analysis described as bellow,

Figure 3. Range of chemical potentials (Pb and I) satisfying eqs 3 and 4. The only colored region is the thermodynamically stable for equilibrium growth conditions of MAPbI3. Adapted with permission from ref 41. Copyright 2014 American Chemical Society.

Ω(ΔμPb , ΔμI , Δμ MAPbI ) 3

metal ≈ Etot[Pbα Iβ(MAPbI3)γ ] − αμPb −

tetragonal − γμ MAPbI − αΔμPb − β ΔμI − γ Δμ MAPbI

3

3

metal ΔμPb = μPb − μPb ,

ΔμI = μI −

tetragonal Δμ MAPbI = μ MAPbI − μ MAPbI 3

3

3

recent studies.38,58 The right, upper, right upper, and left lower limits represent the metal Pb-, gas I2-, PbI2-, and MAI-rich conditions, respectively. Outside this region, MAPbI3 will decompose into each ingredient. Thus, the thermodynamically stable range is fairly narrow, which indicates that MAPbI3 crystals are easy to decompose into PbI2 and MAI (the dissociation energy is ∼0.1 eV/formula) or other elements. Then, the grand potentials Ω of the selected terminations are evaluated from DFT calculations, that is, total energies of relaxed slab structures. Figure 4 shows surface termination diagrams of the four surfaces. All diagrams have large surface regions of (MAPbI3)γ and (PbI2)α(MAPbI3)γ depicted as the dark and light blue areas, respectively. The former and latter terminations can be categorized as vacant- and flat-type terminations, respectively, excluding the (100) surface. We call them stable vacant and PbI2-rich flat in this Account. The thermodynamically stable range is placed mostly on the stablevacant regions, which indicates predominance of this type of surface in PSCs. In addition, the flat termination regions locate close to the thermodynamically stable range. Actually, differences among the grand potentials per unit area of the main two terminations on the (110), (001), and (101) surfaces are small, for which values are within 0.3 eV/nm2. Consequently, the stable-vacant and PbI2-rich-flat terminations can coexist on many MAPbI3 surfaces, and the latter is expected to become

β gas μ 2 I2 (1)

1 gas μ , 2 I2 (2)

Here Etot[PbαIβ(MAPbI3)γ] is total energy of surface slab composed of α Pb atoms, β I atoms, and γ MAPbI3 complexes, and μi (i = Pb, I, MAPbI3) are chemical potentials of individual constitute elements. Reference chemical potentials in Pb metal, I2 gas, and MAPbI3 tetragonal phases are represented as μmetal Pb , tetragonal μgas I2 , and μMAPbI3 , respectively. All of the components are assigned by DFT total energies per unit. Variations of the chemical potentials from those references are Δμi, which define the environmental conditions. Following well developed DFT studies applying with surface stabilities, eq 1 have already ignored entropy terms.55,56 It is assumed that the system is always under equilibrium with MAPbI3, that is, ΔμMAPbI3 = 0. Consider phase-balance conditions under Pb metal, I2 gas, MAI, and PbI2, thermodynamically stable ranges for equilibrium growth of MAPbI3 is described as C

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Table 1. Calculated Energy Gaps of the Stable-Vacant and PbI2-Rich-Flat Terminations (in eV)a (110) (001) (100) (101)

stable vacant

PbI2-rich flat

1.91 1.83 1.87 1.87

1.63 1.56 1.77 1.77

a

Reprinted with permission from ref 41. Copyright 2014 American Chemical Society.

surfaces could be regarded as origin of these alterations. Generally, such structures with lower symmetry show larger energy gaps. In MAPbI3 crystal, observed gaps increase after the transition from the tetragonal to orthorhombic phase, which can be understood by the same explanation about low symmetry. Actually, our first-principles calculations confirmed that the structural change of tetragonal to orthorhombic phase increases the energy gap from 1.64 to 1.70 eV.41 Similar phenomena were discussed about three- and two-dimensional MAPbI3 crystals59 and also band gap engineering through structural modifications.60,61 We should be careful about the energy gaps in Table 1 because these values can have slab size dependences. In the stable-vacant cases, it is difficult to obtain the converged energy gaps compared with the bulk value, because of the distortions of PbI6 octahedron even in the middle region. In the PbI2-richflat terminations, in contrast, the distortions are rather small at the subsurface region, and the value of energy gap show almost no change even with a small slab. However, the calculations of the larger slabs confirmed that our discussions on the electronic properties below are insensitive to the size of calculated slab and broadly applicable.41

Figure 4. Phase diagrams for surface terminations at various chemical conditions: (a) (110), (b) (001), (c) (100), and (d) (101) surfaces. Regions of the stable-vacant and PbI2-rich-flat terminations are marked as dark and light blue colors, respectively. Relaxed surface structures of two terminations are illustrated in the right panels. Adapted with permission from ref 41. Copyright 2014 American Chemical Society.



ELECTRONIC STRUCTURES AND SURFACE STATES For simplicity, we focus on the two main phases on the dominant (110) and (001) surfaces, that is, the stable-vacant and the PbI2-rich-flat terminations. Figure 5 depicts projected density of states (PDOSs) calculated on the selected four surfaces. Their valence and conduction bands consist mostly of I 5p and Pb 6p orbitals, respectively, which is almost the same as bulk phase.59 In addition, PDOSs from MA molecules locate far from their energy gap.43 A number of previous DFT calculations expected splitting of the Pb 6p bands by including a spin−orbit coupling (SOC).35−37,62 Thus, the energy gaps of surface slab could decrease by inclusion of these relativistic effects, which will be discussed later. We emphasize that deep midgap states cannot arise on all of the PDOSs in Figure 5. This means that electron−hole recombination hardly occurs even at surfaces. Therefore, the extreamly large diffusion lengths of the photoexcited carriers in organolead halide perovskites can be realized.24−30 Here, we introduce properties of surface states on the (110) vacant and flat terminations. Similar to the bulk phase, the surface slabs of both terminations have strong antibonding character between I 5p and Pb 6s orbitals. The antibonding coupling of the vacant terminations is weakened because of the absence of PbIx polyhedrons at the outermost layer. Thus, their surface states, which are located at 0.2 eV below the HOMO level, show lower energy than the top of the valence bands. Consequently, the strong antibonding character within I 5p bonds could be essential for the long hole diffusion lengths.

dominant under the PbI2-rich growth condition. These discussions can be applied and qualitatively unchanged for other calculation conditions with larger slabs and a different functional, e.g. including van der Waals interactions.41 The relaxed structures of the main two terminations, that is, stable-vacant and PbI2-rich-flat, are also shown in Figure 4. The (110)-(PbI5)(PbI3) termination forms PbI4 tetrahedrons at the outermost layers, although other terminations keep their initial structures of outermost PbIx polyhedrons. In the stable-vacant terminations, the lack of surface layer leads to large symmetry breaking, especially on (110) and (001) surfaces. At the same time, a wide range of distortions of PbI6 octahedron in the bulk part appears. In contrast, all of the PbI2-rich-flat terminations show little structural deviations from the crystal configuration of tetragonal MAPbI3. We checked the values of grand potentials employing large sized surface slabs of the (110) vacant and flat terminations in ref 41. The difference in the grand potential per area between the two terminations is almost converged to ca. 0.3 eV/nm2 at the present slab. Thus, the phase diagrams presented here qualitatively hold in larger slabs. The difference of the surface terminations could also change their energy gaps, as listed in Table 1. In comparison with the PbI2-rich-flat, the energy gaps of stable-vacant types are large (difference among them are about 0.2 eV), which is attributed to the distortions of PbI6 octahedron, for example, increase of Pb−I distances. Blank of the PbI4 polyhedrons on stable-vacant D

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

Figure 6. PDOS and LUMOs of the PbI2-rich-flat terminations on the (a) (110) and (b) (001) surfaces, including SOC. Adapted with permission from ref 41. Copyright 2014 American Chemical Society.

not change, the levels of the LUMOs decrease about 1 eV in both surfaces. Compared with the LUMOs without SOC, charge distributions of LUMOs with SOC spread and delocalize, for which results are the same as the previous reports for bulk phase.35−37,62 Because the large SOC effects come from the Pb 6p character, the energy shifts of conduction bands can be expected all of surfaces. Note that the orbital characters around the band gap do not alter even in use of SOC (and other functionals) in principle. Thus, the electronic properties discussed in this paper will hold.



DISCUSSIONS AND FUTURE PROSPECTS The present results suggest that the stable-vacant terminations are mainly formed at entire region of MAPbI3 growth conditions. The one-step methods could produce this type of surfaces. We elucidated that several surface structures have no states in midgap region. At nearly the same time, Buin et al. evaluated DOS of MAI and PbI2 terminated surfaces and found no states in its gap.40 These results revealed that electron−hole recombination in organolead halide perovskites is forbidden even at surfaces, the observed long lifetimes of the photoexcited carriers are well explained in the same way as bulk defects.38−41 However, the stable-vacant surfaces have disadvantages for the ability of hole transfer. In contrast, the PbI2-rich-flat surfaces, which could be formed by the two-step method because of PbI2-rich condition, enable the hole transfer effectively. We emphasize that the stoichiometry of the flat termination slabs are consistent with the small composition ratios of the I atom observed only in the two-step method.50 Unfortunately, it is expected that coverage ratio of the flat terminations might be small, because, in the thermodynamically allowed range, the most stable surfaces is the stable-vacant terminations. In addition, a previous study reported by Dong et al. indicates that electron trap states appear on Pb dimer adsorbed PbI2terminated (001) surface.63 Therefore, we can expect that realization of larger area of the PbI2-rich-flat surfaces significantly contributes to an increase in IPCE and further improvements of PSC for photovoltaic applications.

Figure 5. PDOSs of selected MAPbI3 surfaces. Black, green, red, orange, and blue lines indicate PDOSs of total, I 5s, I 5p, Pb 6s, and Pb 6p, respectively. The energy origins are set to the HOMO levels. Corresponding charge distributions of the HOMOs are also illustrated. Adapted with permission from ref 41. Copyright 2014 American Chemical Society.

Figure 5 also depicts charge distributions of the highest occupied molecular orbitals (HOMOs) corresponding to each PDOS. The HOMOs of the vacant and flat terminations are distributed inside of the whole slab and located at surface layers, respectively. Actually, these states rather localized on flat (110) surfaces possibly become recombination centers. However, the HOMOs around the surfaces can facilitate the hole transfer from MAPbI3 to the adjacent HTMs. This in fact accounts for the observed long and fast carrier diffusion. On the other hand, charge transfer through the HOMOs of the vacant surfaces is not effective because coupling between them and HTMs can be expected to be small. Because the energy levels of the flat terminations, which surface states are located just above the top of the valence bands, indicate that they can collect holes generated under their levels and energy loss through the charge transfer is negligible. Now, the SOC effects are introduced. Figure 6 shows PDOSs and the lowest unoccupied molecular orbitals (LUMOs) of the PbI2-rich-flat terminations on (110) and (001) planes. While the valence bands approximately do E

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Accounts of Chemical Research



Although the surface properties of organolead halide perovskites are important for prevailing carrier recombination and realizing a smooth carrier transport, interface properties between the perovskite material and the carrier transport layers are also essential for efficient charge transfer processes. We consider that first-principles investigations of these interfaces, for example, studies by De Angelis and co-workers,19,64,65 are necessary for predicting the band alignment. Carrier transportations from perovskites to adjacent layers are induced by the band-offset near the interfaces. The band alignments between the perovskites and TiO2 are favorable for electron transportation. First-principles calculations could find out the optimized materials at interfaces through changing carrier transporting materials, metal/halide/molecule ratios and components of perovskite absorbers. Here, we should point out instabilities of the MAPbI3/HTM interfaces, especially HTM containing spiro-OMeTAD. There are many suggestions that the HTM-free or alternative spiro-OMeTAD layer can effectively improve the cell stability.14,66 As the enhanced durability of the PSCs could be attributed to the avoidance of the use of deliquescent additives, finding the alternative of spiro-OMeTAD or suppressing the degradation by water is essential for the cell durability. Surprisingly, Mosconi et al. have pointed out the termination dependence also from the perspective of the degradation. They conducted DFT molecular dynamics simulation at MAPbI3/water interfaces. Their simulations indicate that PbI2-rich-flat terminations act as a protective layer against the water degradation, but PbI2 defects, the structure is almost the same as the present stable vacant termination, trigger the facile solvation.67 Furthermore, firstprinciple treatment of charge transfer process and chargeseparation dynamics at the interfaces is essential for the further development of PSCs because a detailed mechanism of electron/hole transfer within this photovoltaic system has a lack of understanding. For example, the role of Al2O3 is under the discussion.4,15,49 Comparing with the calculation and the experimental results, e.g. charge transfer rates and passivation effects could help the understanding of a charge-separation dynamics at interface between the absorber and carrier transport layers. We believe that these investigations provide more detailed insight of interface roles and future direction for the PSC developments.



Article

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Jun Haruyama was born 1985 in Tokyo, Japan. He received his Ph.D. in physics from Tokyo University of Science (2013). He is a postdoctoral fellow at National Institute for Materials Science (NIMS), Japan. His main research concerns the application of firstprinciples calculation to solid-state interfaces, with a particular emphasis on perovskite solar cells and all-solid-state Li-ion batteries. Keitaro Sodeyama was born 1975 in Tokyo, Japan. He received his Ph.D. in chemistry from Waseda University (2005). He is a project researcher at Kyoto University, Japan. His main research interest is first-principles MD simulation of solid/liquid interface in solar cells and batteries. Liyuan Han was born 1956 in Shanghai, China. He received his Ph.D. in applied chemistry from University of Osaka Prefecture (1988). He is the director of the Photovoltaic Materials Unit, NIMS. His current research focuses on developments of higly efficient perovskite/dyesensitized solar cells. He is an inventor with over 100 patents and an author in the same number of scientific publications. Yoshitaka Tateyama was born 1970 in Aomori, Japan. He received his Ph.D. in physics from The University of Tokyo (1998). He is a group leader in NIMS. His main research interest is first-principles based free-energy investigation on redox reactions at solid/liquid and solid/ solid interfaces.



ACKNOWLEDGMENTS The calculations were carried out on the supercomputers in NIMS and The University of Tokyo as well as the supercomputers in Kyushu University through the HPCI Systems Research Projects (Proposal Nos. hp140179, hp140110, hp140232, hp150055, and hp150068).



REFERENCES

(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, 6050−6051. (2) Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G. 6.5% Efficient Perovskite Quantum-Dot-Sensitized by Lead-Halide Compounds. Nanoscale 2011, 3, 4088−4093. (3) 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.; Grätzel, M.; Park, N.-G. Lead Iodide Perovskite Sensitized AllSolid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (4) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Efficient Hybrid Solar Cells Based on Meso−Superstructured Organometal Halide Perovskite. Science 2012, 338, 643−647. (5) Burschka, J.; Pellet, N.; Moon, S. J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Grätzel, M. Sequential Deposition as a Route to High-Performance Perovskite-Sensitized Solar Cells. Nature 2013, 499, 316−319. (6) Liu, M.; Johnston, M. B.; Snaith, H. J. Efficient Planar Heterojunction Perovskite Solar Cells by Vapour Deposition. Nature 2013, 501, 395−398.

CONCLUSIONS

The structural stability at various chemical environments was investigated on the (110), (001), (100), and (101) surfaces of tetragonal MAPbI3 within DFT calculations. We found the termination dependence of the structural stability and electronic states. The vacant terminations are more probable than the flat one on many surfaces, but both terminations can coexist particularly on (110) and (001) surfaces. Electronic structures of the vacant and flat terminations on the two surfaces are almost the same as that in the bulk MAPbI3. Thus, these surfaces can realize the long carrier diffusion lengths observed in experiments. In addition, the flat terminations on (110) and (001) surfaces have surface states, for which energy levels are located just above the bulk valence band. Thus, they could be efficient intermediates for hole transfer. As a result, increasing coverage of the flat terminations will be of great benefit for the further improvement of PSC performances. F

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research (7) Liu, D.; Kelly, T. L. Perovskite Solar Cells with a Planar Heterojunction Structure Prepared Using Room-Temperature Solution Processing Techniques. Nat. Photonics 2014, 8, 133−138. (8) 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. (9) Zhou, H.; Chen, Q.; Li, G.; Luo, S.; Song, T.-B.; Duan, H.-S.; Hong, Z.; You, J.; Liu, Y.; Yang, Y. Interface Engineering of Highly Efficient Perovskite Solar Cells. Science 2014, 345, 542−546. (10) Im, J.-H.; Jang, I.-H.; Pellet, N.; Grätzel, M.; Park, N.-G. Growth of CH3NH3PbI3 Cuboids with Controlled Size for High-Efficiency Perovskite Solar Cells. Nat. Nanotechnol. 2014, 9, 927−932. (11) 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. (12) Best Reserch-Cell Efficiencies; National Renewable Energy Laboratory: Washington, DC; http://www.nrel.gov/ncpv/images/ efficiency_chart.jpg (accessed August 9, 2015). (13) Wu, Y.; Islam, A.; Yang, X.; Qin, C.; Liu, J.; Zhang, K.; Peng, W.; Han, L. Retarding the Crystallization of PbI2 for highly reproducible Planar-Structured Perovskite Solar Cells via Sequential Deposition. Energy Environ. Sci. 2014, 7, 2934−2938. (14) Liu, J.; Wu, Y.; Qin, C.; Yang, X.; Yasuda, T.; Islam, A.; Zhang, K.; Peng, W.; Chen, W.; Han, L. A Dopant-Free Hole-Transporting Material for Efficient and Stable Perovskite Solar Cells. Energy Environ. Sci. 2014, 7, 2963−2967. (15) Chen, W.; Wu, Y.; Liu, J.; Qin, C.; Yang, X.; Islam, A.; Cheng, Y.-B.; Han, L. Hybrid Interfacial Layer Leads to Solid Peroformance Improvement of Inverted Perovskite Solar Cells. Energy Environ. Sci. 2015, 8, 629−640. (16) Liu, J.; Shirai, Y.; Yang, X.; Yue, Y.; Chen, W.; Wu, Y.; Islam, A.; Han, L. High-Quality Mixed-Organic-Cation Perovskites from a PhasePure Non-Stoichiometric Intermediate (FAI)1‑x-PbI2 for Solar Cells. Adv. Mater. 2015, 27, 4918−4923. (17) O’Regan, B.; Grätzel, M. A Low-Cost, High-Efficiency Solar Cell Based on Dye-Sensitized Colloidal TiO2 Films. Nature 1991, 353, 737−740. (18) Zhang, S.; Yang, X.; Numata, Y.; Han, L. Highly Efficient DyeSensitized Solar Cells: Progress and Future Challenges. Energy Environ. Sci. 2013, 6, 1443−1464. (19) De Angelis, F. Modeling Materials and Processes in Hybrid/ Organic Photovoltaics: From Dye-Sensitized to Perovskite Solar Cells. Acc. Chem. Res. 2014, 47, 3349−3360. (20) Tateyama, Y.; Sumita, M.; Ootani, Y.; Aikawa, K.; Jono, R.; Han, L.; Sodeyama, K. Acetonitrile Solution Effect on Ru N749 Dye Adsorption and Excitation at TiO2 Anatase Interface. J. Phys. Chem. C 2014, 118, 16863−16871. (21) Snaith, H. J. Perovskites: The Emergence of a New Era for LowCost, High-Efficiency Solar Cells. J. Phys. Chem. Lett. 2013, 4, 3623− 3630. (22) Park, N.-G. Organometal Perovskite Light Absorbers Toward a 20% Efficiency Low-Cost Solid-State Mesoscopic Solar Cell. J. Phys. Chem. Lett. 2013, 4, 2423−2429. (23) De Wolf, S.; Holovsky, J.; Moon, S.-J.; Löper, P.; Niesen, B.; Ledinsky, M.; Haug, F.-J.; Yum, J.-H.; Ballif, C. Organometallic Halide Perovskites: Sharp Optical Absorption Edge and Its Relation to Photovoltaic Performance. J. Phys. Chem. Lett. 2014, 5, 1035−1039. (24) 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, 341− 344. (25) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- and Hole-Transport Lengths in Organic-Inorganic CH3NH3PbI3. Science 2013, 342, 344−347. (26) Eperon, G. E.; Stranks, S. D.; Menelaou, C.; Johnston, M. B.; Herz, L. M.; Snaith, H. J. Formamidinium Lead Trihalide: a Broadly

Tunable Perovskite for Efficient Planar Heterojunction Solar Cells. Energy Environ. Sci. 2014, 7, 982−988. (27) Dong, Q.; Fang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.; Huang, J. Electron-Hole Diffusion Lengths > 175 μm in SolutionGrown CH3NH3PbI3 Single Crystals. Science 2015, 347, 967−970. (28) Ponseca, C. S., Jr.; Savenije, T. J.; Abdellah, M.; Zheng, K.; Yartsev, A.; Pascher, T.; Harlang, T.; Chabera, P.; Pullerits, T.; Stepanov, A.; Wolf, J.-P.; Sundströ m, V. Organometal Halide Perovskite Solar Cell Materials Rationalized: Ultrafast Charge Generation, High and Microsecond-Long Balanced Mobilities, and Slow Recombination. J. Am. Chem. Soc. 2014, 136, 5189−5192. (29) Savenije, T. J.; Ponseca, C. S., Jr.; Kunneman, L.; Abdellah, M.; Zheng, K.; Tian, Y.; Zhu, Q.; Canton, S. E.; Scheblykin, I. G.; Pullerits, T.; Yartsev, A.; Sundströ m, V. Thermally Activated Exciton Dissociation and Recombination Control the Carrier Dynmics in Organolead Halide Perovskite. J. Phys. Chem. Lett. 2014, 5, 2189− 2194. (30) Oga, H.; Saeki, A.; Ogomi, Y.; Hayase, S.; Seki, S. Improved Understanding of the Electronic and Energetic Landscapes of Perovskite Solar Cells: High Local Charge Carrier Mobility, Reduced Recombination, and Extremely Shallow Traps. J. Am. Chem. Soc. 2014, 136, 13818−13825. (31) Yamada, Y.; Nakamura, T.; Endo, M.; Wakamiya, A.; Kanemitsu, Y. Photocarrier Recombination Dynamics in Perovskite CH3NH3PbI3 for Solar Cell Applications. J. Am. Chem. Soc. 2014, 136, 11610− 11613. (32) Manser, J. S.; Kamat, P. V. Band Filling with Free Charge Carriers in Organometal Halide Perovskites. Nat. Photonics 2014, 8, 737−743. (33) Yin, W.-J.; Yang, J.-H.; Kang, J.; Yan, Y.; Wei, S.-H. Halide Perovskite Materials for Solar Cells: a Theoretical Review. J. Mater. Chem. A 2015, 3, 8926−8942. (34) Yin, W.-J.; Shi, T.; Yan, Y. Unique Properties of Halide Perovskites as Possible Origins of the Superior Solar Cell Performance. Adv. Mater. 2014, 26, 4653−4658. (35) Giorgi, G.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Small Photocarrier Effective Masses Featuring Ambipolar Transport in Methylammonium Lead Iodide Perovskite: A Density Functional Analysis. J. Phys. Chem. Lett. 2013, 4, 4213−4216. (36) Umari, P.; Mosconi, E.; De Angelis, F. Relativistic GW Calculations on CH3NH3PbI3 and CH3NH3SnI3 Perovskites for Solar Cell Applications. Sci. Rep. 2014, 4, 4467. (37) Feng, J.; Xiao, B. Crystal Structures, Optical Properties, and Effective Mass Tensors of CH3NH3PbX3 (X = I and Br) Phases Predicted from HSE06. J. Phys. Chem. Lett. 2014, 5, 1278−1282. (38) Yin, W.-J.; Shi, T.; Yan, Y. Unusual Defect Physics in CH3NH3PbI3 Perovskite Solar Cell Absorber. Appl. Phys. Lett. 2014, 104, 063903. (39) Kim, J.; Lee, S.-H.; Lee, J. H.; Hong, K.-H. The Role of Intrinsic Defects in Metylammonium Lead Iodide Perovskite. J. Phys. Chem. Lett. 2014, 5, 1312−1317. (40) Buin, A.; Pietsch, P.; Xu, J.; Voznyy, O.; Ip, A. H.; Comin, R.; Sargent, E. H. Materials Processing Routes to Trap-Free Halide Perovskites. Nano Lett. 2014, 14, 6281−6286. (41) Haruyama, J.; Sodeyama, K.; Han, L.; Tateyama, Y. Termination Dependence of Tetragonal CH3NH3PbI3 Surfaces for Perovskite Solar Cells. J. Phys. Chem. Lett. 2014, 5, 2903−2909. (42) Etgar, L.; Gao, P.; Xue, Z.; Peng, Q.; Chandiran, A. K.; Liu, B.; Nazeeruddin, M. K.; Grätzel, M. Mesoscopic CH3NH3PbI3/TiO2 Heterojunction Solar Cells. J. Am. Chem. Soc. 2012, 134, 17396− 17399. (43) Baikie, T.; Fang, Y.; Kadro, J. M.; Schreyer, M.; Wei, F.; Mhaisalkar, S. G.; Grätzel, M.; White, T. J. Synthesis and Crystal Chemistry of the Hybrid Perovskite (CH3NH3)PbI3 for Solid-State Sensitized Solar Cell Applications. J. Mater. Chem. A 2013, 1, 5628− 5641. (44) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Semiconducting Tin and Lead Iodide Perovskites with Organic G

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX

Article

Accounts of Chemical Research

(65) Mosconi, E.; Ronca, E.; De Angelis, F. First-Principles Investigation of the TiO2/Organohalide Perovskites Interface: The Role of Interfacial Chlorine. J. Phys. Chem. Lett. 2014, 5, 2619−2625. (66) Mei, A.; Li, X.; Liu, L.; Ku, Z.; Liu, T.; Rong, Y.; Xu, M.; Hu, M.; Chen, J.; Yang, Y.; Grätzel, M.; Han, H. A Hole-Conductor-Free, Fully Printable Mesoscopic Perovskite Solar Cell with High Stability. Science 2014, 345, 295−298. (67) Mosconi, E.; Azpiroz, J. M.; De Angelis, F. Ab Initio Molecular Dynamics Simulations of Methylammonium Lead Iodide Perovskite Degradation by Water. Chem. Mater. 2015, 27, 4885−4892.

Cations: Phase Transitions, High Mobilities, and Near-Infrared Photoluminescent Properties. Inorg. Chem. 2013, 52, 9019−9038. (45) Kawamura, Y.; Mashiyama, H.; Hasebe, K. Structural Study on Cubic−Tetragonal Transition of CH3NH3PbI3. J. Phys. Soc. Jpn. 2002, 71, 1694−1697. (46) Momma, K.; Izumi, F. VESTA: a Three-Dimensional Visualization System for Electronic and Structural Analysis. J. Appl. Crystallogr. 2008, 41, 653−658. (47) 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.; Grät zel, M.; Seok, S. I. Efficient Inorganic−Organic Hybrid Heterojunction Solar Cells Containing Perovskite Compound and Polymeric Hole Conductors. Nat. Photonics 2013, 7, 486−491. (48) Starr, D. E.; Sadoughi, G.; Handick, E.; Wilks, R. G.; Alsmeier, J. H.; Köhler, L.; Gorgoi, M.; Snaith, H. J.; Bär, M. Direct Observation of an Inhomogeneous Chlorine Distribution in CH3NH3PbI3‑xClx Layers: Surface Depletion and Interface Enrichment. Energy Environ. Sci. 2015, 8, 1609−1615. (49) Grätzel, M. The Light and Shade of Perovskite Solar Cells. Nat. Mater. 2014, 13, 838−842. (50) Lindblad, R.; Bi, D.; Park, B.-W.; Oscarsson, J.; Gorgoi, M.; Siegbahn, H.; Odelius, M.; Johansson, E. M. J.; Rensmo, H. Electronic Structure of TiO2/CH3NH3PbI3 Perovskite Solar Cell Interfaces. J. Phys. Chem. Lett. 2014, 5, 648−653. (51) Mosconi, E.; Amat, A.; Nazeeruddin, M. K.; Grätzel, M.; De Angelis, F. First-Principles Modeling of Mixed Halide Organometal Perovskites for Photovoltaic Applications. J. Phys. Chem. C 2013, 117, 13902−13913. (52) Giorgi, G.; Fujisawa, J.-I.; Segawa, H.; Yamashita, K. Organic− Inorganic Hybrid Lead Iodide Perovskite Featuring Zero Dipole Moment Guanidinum Cations: A Theoretical Analysis. J. Phys. Chem. C 2015, 119, 4694−4701. (53) Azpiroz, J. M.; Mosconi, E.; Bisquert, J.; De Angelis, F. Defect Migration in Methylammonium Lead Iodide and Its Role in Perovskite Solar Cell Operation. Energy Environ. Sci. 2015, 8, 2118−2127. (54) Haruyama, J.; Sodeyama, K.; Han, L.; Tateyama, Y. FirstPrinciples Study of Ion Diffusion in Perovskite Solar Cell Sensitizers. J. Am. Chem. Soc. 2015, 137, 10048−10051. (55) Qian, G.-X.; Martin, R. M.; Chadi, D. J. Stoichiometry and Surface Reconstruction: An Ab Initio Study of GaAs(100) Surfaces. Phys. Rev. Lett. 1988, 60, 1962−1965. (56) Sanna, S.; Schmidt, W. G. Lithium Niobate X-Cut, Y-Cut, and Z-Cut Surfaces from Ab Initio Theory. Phys. Rev. B: Condens. Matter Mater. Phys. 2010, 81, 214116. (57) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (58) Agiorgousis, M. L.; Sun, Y.-Y.; Zeng, H.; Zhang, S. Strong Covalency-Induced Recombination Centers in Perovskite Solar Cell Material CH3NH3PbI3. J. Am. Chem. Soc. 2014, 136, 14570−14575. (59) Umebayashi, T.; Asai, K.; Kondo, T.; Nakao, A. Electronic Structures of Lead Iodide Based Low-Dimensional Crystals. Phys. Rev. B: Condens. Matter Mater. Phys. 2003, 67, 155405. (60) Filip, M. R.; Eperon, G. E.; Snaith, H. J.; Giustino, F. Steric Engineering of Metal-Halide Perovskites with Tunable Optical Band Gaps. Nat. Commun. 2014, 5, 5757. (61) Kim, J.; Lee, S.-C.; Lee, S.-H.; Hong, K.-H. Importance of Orbital Interactions in Determining Electronic Band Structures of Organo-Lead Iodide. J. Phys. Chem. C 2015, 119, 4627−4634. (62) Even, J.; Pedesseau, L.; Jancu, J.-M.; Katan, C. Importance of Spin−Orbit Coupling in Hybrid Organic/Inorganic Perovskites for Photovoltaic Applications. J. Phys. Chem. Lett. 2013, 4, 2999−3005. (63) Dong, R.; Fang, Y.; Chae, J.; Dai, J.; Xiao, Z.; Dong, Q.; Yuan, Y.; Centrone, A.; Zeng, X. C.; Huang, J. High-Grain and Low-DrivingVoltage Photodetectors Based on Organolead Triiodide Perovskites. Adv. Mater. 2015, 27, 1912−1918. (64) Roiati, V.; Mosconi, E.; Listorti, A.; Colella, S.; Gigli, G.; De Angelis, F. Stark Effect in Perovskite/TiO2 Solar Cells: Evidence of Local Interfacial Order. Nano Lett. 2014, 14, 2168−2174. H

DOI: 10.1021/acs.accounts.5b00452 Acc. Chem. Res. XXXX, XXX, XXX−XXX