Thermodynamically Stable Orthorhombic Î³-CsPbI3 Thin Films for High

Subscriber access provided by University of South Dakota

Article

Thermodynamically Stable Orthorhombic #-CsPbI3 Thin Films for High-Performance Photovoltaics Boya Zhao, Shifeng Jin, Sheng Huang, Ning Liu, Jing-Yuan Ma, Ding-Jiang Xue, Qiwei Han, Jie Ding, Qian-Qing Ge, Yaqing Feng, and Jin-Song Hu J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 28 Aug 2018 Downloaded from http://pubs.acs.org on August 28, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Thermodynamically Stable Orthorhombic ‐CsPbI3 Thin Films for High‐Performance Photovoltaics Boya Zhao,†,‡ Shi-Feng Jin,§ Sheng Huang,|| Ning Liu,§ Jing-Yuan Ma,†,# Ding-Jiang Xue,†,# Qiwei Han,¶ Jie Ding,†,# Qian-Qing Ge,†,# Yaqing Feng,*,‡ and Jin-Song Hu*,†,# †National

Research Center for Molecular Sciences, Key Laboratory of Molecular Nanostructure and Nanotechnology, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China ‡School of Chemical Engineering and Technology, Tianjin University, Tianjin 300350, China §Institute of Physics, Chinese Academy of Sciences, Beijing 100190, China ||Beijing Key Laboratory of Nanophotonics and Ultrafine Optoelectronic Systems, School of Materials Science and Engineering, Beijing Institute of Technology, Beijing 100081, China #University of Chinese Academy of Sciences, Beijing 100049, China ¶Department of Chemistry, Duke University, Durham, NC 27708, USA. ABSTRACT: All-inorganic lead halide perovskites demonstrate improved thermal stability over the organic-inorganic halide perovskites, but the cubic -CsPbI3 with the most appropriate bandgap for light harvesting is not structurally stable at room temperature and spontaneously transforms into the undesired orthorhombic -CsPbI3. Here, we present a new member of black-phase thin films of all-inorganic perovskites for high-efficiency photovoltaics, the orthorhombic -CsPbI3 thin films with intrinsic thermodynamic stability and ideal electronic structure. Exempt from introducing organic ligands or incorporating mixed cations/anions into the crystal lattice, we stabilize the -CsPbI3 thin films by a simple solution process in which a small amount of H2O manipulates the size-dependent phase formation through a proton transfer reaction. Theoretical calculations coupled with experiments show that -CsPbI3 with a lower surface free energy becomes thermodynamically preferred over -CsPbI3 at surface areas greater than 8,600 m2/mol and exhibits comparable optoelectronic properties to -CsPbI3. Consequently, -CsPbI3-based solar cells display a highly reproducible efficiency of 11.3%, among the highest records for CsPbI3 thin-film solar cells, with robust stability in ambient atmosphere for months and continuous operating conditions for hours. Our study provides a novel and fundamental perspective to overcome the Achilles’ heel of the inorganic lead iodide perovskite and opens it up for high-performance optoelectronic devices.

been made through alloying with other cations or anions into the perovskite lattice.11-13 For instance, partial substitution of iodide with bromide has been widely reported to improve the stability of the perovskite phase, but compromise on light harvesting with increased bandgap.9, 14 Encouragingly, interface modification and optimization have recently been demonstrated to notably enhance the performance of CsPbI2Br devices by improving carrier collection and conduction.15-19 Additionally, it has been shown that with the assistance of organic ligands, the CsPbI3 colloidal quantum dots as well as the bulk thin films can be stabilized in the cubic structure at room temperature.13, 20-24 However, the presence of organic part may render the perovskite vulnerable to heat again, trading off thermal stability for structural stability. Another approach to obtain -CsPbI3 is to introduce hydroiodic acid (HI) into the precursor solution, which reduces the phase transition temperature in large measure.25-27 A review of the up-to-date reports on

■ INTRODUCTION Organic-inorganic halide perovskite solar cells (PSCs) have achieved remarkable advances with the power conversion efficiency (PCE) soaring above 22% in less than a decade.16 However, the thermal instability of organometal halide perovskites has become a roadblock towards commercialization due to the volatile nature of the organic component such as methylammonium (CH3NH3) and formamidinium [HC(NH2)2].7-8 Alternatively, all-inorganic lead halide perovskites (CsPbX3) are much desired owing to the improved composition stability under thermal stress,9 with the cubic -CsPbI3 (black phase) exhibiting the most suitable bandgap (Eg  ~1.7 eV) for photovoltaic applications among all CsPbX3 candidates.10 Unfortunately, -CsPbI3 is structurally unstable below 320 ℃ and undergoes a spontaneous phase transition to orthorhombic -CsPbI3 (yellow phase, Eg  ~2.8 eV) with a poor response to the solar spectrum.10-11 Attempts to stabilize black-phase CsPbI3 at room temperature have

1

ACS Paragon Plus Environment

Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 12

Figure 1. Structural characterization and phase stability of CsPbI3 perovskite. (a) XRD patterns of CsPbI3 perovskite films prepared with varying H2O concentrations (x), distinguished from the standard pattern of -CsPbI3. (b) Rietveld refinement of the x = 0.02 film XRD pattern, revealing pure orthorhombic -CsPbI3 with the Pbnm space group. (c) Crystal structure of -phase CsPbI3 viewed along the [110] and [001] directions, respectively. Phase stability of (d) orthorhombic -CsPbI3 and (e) tetragonal -CsPbI3 stored in ambient conditions for one month as displayed by XRD and photographs. The asterisks in the XRD patterns denote the diffraction peaks from the FTO glass.

inorganic PSCs is given in Table S1. Although -CsPbI3 could be stabilized below the phase transition point (320 ℃ ) by the above methods, the unwanted nonperovskite  phase, nevertheless, is thermodynamically preferred in ambient conditions. Thus, it is an imperative for a better understanding of the thermodynamic behaviors of CsPbI3 polymorphs. More importantly, blackphase thin films with intrinsic structural stability, without compromising on heat resistance and electronic structure appositeness, are highly desired for efficient solar cells. Here, we present a new member of black-phase thin films, the orthorhombic -phase thin films of CsPbI3 perovskite, for high-efficiency and stable photovoltaics. Without introducing organic ligands or incorporating mixed cations/anions into the crystal lattice, we stabilize the -CsPbI3 thin films at room temperature by adding a small amount of H2O to the CsPbI3 precursor solution via a facile one-step deposition method. It is discovered that H2O molecules delicately tune the equilibrium of the proton transfer reaction in the precursor solution and thus manipulate the thermodynamically favored phase through tailoring the dimensions of the perovskite crystallites. The density functional theory (DFT) calculations demonstrate

that -CsPbI3 has a lower surface free energy than -CsPbI3 and therefore becomes thermodynamically stable at significantly extended surface areas beyond 8,600 m2/mol, as evidenced by the phase durability in ambient air over a month. Furthermore, the perovskite-structured  phase shows ideal electronic structure by DFT calculations and comparable optoelectronic performance to its cubic isomer, proving itself to be a promising candidate for photovoltaic applications. As a proof of principle, PSCs based on the -CsPbI3 absorbers achieved a power conversion efficiency (PCE) as high as 11.3%, without efficiency loss in ambient environment for months and continuous operating conditions for hours. ■ RESULTS AND DISCUSSION Structure Determination of ‐Phase CsPbI3 Thin Films. We prepared a range of precursor solutions by dissolving equimolar CsI and PbI2 in the mixture of DMF and HI with the addition of different concentrations of deionized H2O. Hereafter we will refer to the volume ratio of H2O to DMF as x (x  0, 0.01, 0.02, 0.04) for convenience. The CsPbI3 films were fabricated by a single-step spincoating method, followed by low-temperature annealing at

2


Page 3 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Interestingly, however, we instead obtained a tetragonal  phase (space group P4/mbm, no. 127) after the perovskite film swiftly cooled down to room temperature (Figure S3 and Table S3). Considering that the -CsPbI3 polymorph has been detected at high temperature,25, 31 these observations imply a phase transformation from the cubic phase to its tetragonal isomer upon termination of the heating process, suggesting the latter being more structurally stable in thin films. Very recently, Marronnier et al. also observed - and -CsPbI3 by means of temperature-dependent synchrotron XRD measurement. However, the  and  phase could only be temporarily retained in the bulk solids during the cooling cycle from high temperature (374℃).32 For photovoltaic application and scalable solar-cell fabrication, low-temperature solution-processed thin films with thermodynamic stability are much more favorable and even indispensable. Since the structural stability is a main concern for the inorganic lead iodide perovskites, we then tested the phase sustainability of the perovskite films at room temperature. After storage for weeks in nitrogen atmosphere, traces of the nonperovskite  phase are found except for the CsPbI3 (x  0.02) sample (Figure S4). Further, we exposed the orthorhombic -CsPbI3 film together with its tetragonal  analogue to ambient air (temperature: 25  5℃, relative humidity: 10  5%). As shown in Figure 1d, the black  phase exhibits superior phase stability with the XRD spectra nearly unchanged after one month of storage in ambient conditions. In contrast, the  phase converted to the yellow  phase under the same circumstances (Figure 1e). The above results indicate that in ambient environment the  phase manifest itself as the perovskitestructured CsPbI3 standing stable in bulk thin films compared to the cubic  phase and the tetragonal  phase, consistent with the most recent studies32-33. Similar relationships between the structural stability and the crystallographic symmetry are observed in other fully inorganic perovskites such as CsPbCl3, CsPbBr3 and CsSnI3 as well as the organic-inorganic counterparts.34-38

100℃. As a control group, the perovskite film was deposited from the precursor solution without HI or H2O content and annealed at 335℃. To identify the impact of H2O on the lattice structures of the CsPbI3 films, we performed X-ray diffraction (XRD) measurements at room temperature. In Figure 1a, we show the XRD patterns of the resulting CsPbI3 films prepared from a set of precursors. It is evident that none of the XRD peaks could be assigned to the Bragg reflections expected in cubic symmetry, for major deviations (~ 0.3°) from the standard -CsPbI3 perovskite structure (space group Pm3m, no. 221) are markedly displayed. Moreover, additional peaks are observed at positions between integral values of Miller indices of -CsPbI3, suggesting a lowering of crystal symmetry by titling the [PbI6]4 octahedra that constitute the perovskite lattice. As the H2O content (x) increases to 0.02, the intensity of the diffraction peak at 2θ 28.5° [(004) reflection] rises accompanied by the disappearance of the peak at 25.3° and the emergence of another at 14.1° [(002) reflection]. Most importantly, we managed to index every single peak for the x  0.02 sample to the reflections of a disparate orthorhombic space group, Pbnm (no. 62). Although the x  0.02 film takes on a black appearance (Figure S1) with significant spectral absorption (Figure S2), it is indeed not the cubic  phase but a completely new black phase of CsPbI3 perovskite. By further increasing the H2O concentrations (x), however, some reflections of the Pbnm symmetry such as (002), (103), (211) and (004) disappear in conjunction with some others shifting or turning up, indicating a gradual phase transition from this orthorhombic structure. To determine the crystal structure of the black orthorhombic phase (x 0.02), we performed Rietveld refinement of the corresponding XRD profile. As shown in Figure 1b, no detectable cubic phase, yellow orthorhombic phase (space group Pnma, no. 62) or impurity phase is found. Crystallographic data produced by refinement against the room-temperature XRD data are summarized in Table S2. Consequently, we present a pure orthorhombic Pbnm (no. 62) structure with lattice parameters as a  8.646 Å, b  8.818 Å, c  12.520 Å, along with an angle of 155.6° for the PbIPb bond (Figure 1c). The structural parameters obtained after refinement produced an excellent fit to the diffraction pattern with Rp  6.67%, wRp  8.99%, validating the results. For ease of comparison, we denote this orthorhombic structure as the  phase differing from the nonperovskite orthorhombic  phase 10. It should be noted that the structure retains the three-dimensional (3D) framework with the [PbI6]4 octahedra sharing corners in all three orthogonal directions despite the reduction of the PbIPb bond angle. The octahedral tilting does not undermine the overall 3D architecture of -CsPbI3 given a Goldschmidt tolerance factor (t) of 0.85 in accord with a typical t of 0.8-1.0 for most 3D perovskites28-29 and the mildly bent PbIPb bond with the angle greater than 150°, below which the perovskite structures normally disintegrate.30

Formation of Thermodynamically Stable ‐CsPbI3 Polymorph. The morphology of the perovskite films fabricated with different H2O concentrations was examined by scanning electron microscopy (SEM) and atomic force microscopy (AFM). It is apparent from the top-view SEM images in Figure 2a that the -CsPbI3 film (x  0.02) exhibits a uniform and dense film with continuous coverage on the substrate whereas a poor film morphology with much more pinholes is displayed on the pristine film (x  0) and the film with excessive H2O content (x  0.04). We particularly notice that the perovskite crystallites shrink appreciably from an average size of ~230 nm (x  0) to ~100 nm (x  0.02) as the H2O concentration increases, but expands to a greater size of ~260 nm by further adding the H2O content (x  0.04). As for the control sample prepared without HI in the precursor or the addition of H2O, the film is made up of grains with an estimated size of ~800 nm in average and a maximum over 1 m. These observations were further confirmed by the 3D AFM images (Figure 2b). Moreover, the topology of the films clearly shows that the marked change in crystallite size

In addition, the control sample crystallized at 330℃ was supposed to be the cubic -CsPbI3 in view of the annealing temperature beyond that of the -phase transition.11, 31

3



Page 4 of 12

Figure 2. Morphology of CsPbI3 films and Formation of  phase. (a) Top-view SEM images, (b) 3D AFM images and (c) crosssectional height distributions of CsPbI3 films deposited with different H2O concentrations (x) and without HI as a control sample. Scale bars: 500 nm for (a) and 200 nm for (b). (d) Electrical Conductivity of DMF/HI system with varying H2O content. (e) Size distributions of PbI2 colloids in precursor solutions with variant amounts of H2O content detected by dynamic light scattering. (f) Dependence of average crystallite size on the average colloid size in precursor solutions. (g) Calculated Gibbs free energy G of CsPbI3 polymorphs relative to bulk -CsPbI3 as a function of surface area. The slope of each line gives the surface free energy obtained through molecular dynamics simulations.

and surface coverage lead to distinct flatness and smoothness. We calculated the rootmeansquare (rms) roughness to be 18.54 nm, 12.20 nm, 6.80 nm, 16.05 nm and 30.68 nm for films with the H2O concentration (x) of 0, 0.01, 0.02, 0.04 and the control film, respectively. The CsPbI3 (x  0.02) sample with the smallest crystallites presents an ultrasmooth film with the roughness greatly reduced in comparison with other samples, as is evident from the cross-sectional height distributions of the films (Figure 2c), which is critical for efficient planar heterojunction devices.39 It is worth noting that the pure CsPbI3 crystals with better phase durability were stabilized at diminished dimensions of ~100 nm, which suggests a relation between the crystallite size and the perovskite phase stability. To probe the effect of H2O on the formation of the featured morphology and crystal structures, we turned our

attention to the precursor solutions because of the important role they play in crystallization.39-41 Although hydrogen halides are strong electrolytes in H2O, they show incomplete dissociation in N,N-dimethylformamide (DMF) existing in the form of the molecular complex DMFxHX.4243 The electrical conductivity of the DMF/HI solvent system in Figure 2d displays that with the addition of a small amount of H2O molecules, the specific conductance () of the mixed solvent notably increases. For incomplete ionization, good approximations could be gained from the Ostwald dilution law: 𝐾

(1)

where K is the ionization constant, c is the concentration of HI,  is the equivalent conductance (c) and  is a constant denoting the limiting equivalent conductance (It can be readily inferred from the equation that

4


Page 5 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Scheme for the stabilization of -phase CsPbI3. (i) PbI2 colloids made of lead polyiodide complexes in the precursor solution. (ii) H2O molecules facilitate the ionization of HI from the DMFxHI complexes liberating H and I into free ions, which reduces the size of the colloids to form the corner-sharing [PbI6]4 octahedra intercalated by Cs ions. (iii) Finally, the octahedral framework is converted to the perovskite polymorph with decreased grain size by thermal annealing, stabilizing the orthorhombic  phase of CsPbI3 crystals with surface areas significantly extended.

the ionization of HI is promoted when raising the H2O content. This is because H2O is a strong ionizing medium for electrolytes, which is largely attributed to its high dielectric constant of 82.20 (20℃) versus that of 38.25 (20℃) for DMF. Hence, we propose that a proton transfer reaction is probably induced by the H2O molecules, as described by the following chemical equation: 𝑥I (2) DMF𝑥HI 𝑥H O ⇌ DMF 𝑥H O As a result, the equilibrium of the proton transfer process is delicately tuned by controlling the amount of H2O, which changes the ionization degree of the precursor solutions. Previous reports have demonstrated that the hybrid perovskite precursor solutions consist of not only solvated ions but also colloids made up of lead polyhalide complexes serving as the building blocks for the perovskite crystallites.40 And the solvent acidity and precursor composition remarkably affect the crystallization process where, specifically, an increase in proton activity or I concentration can contribute to smaller lead halide colloids.40-41 In Figure 2e, the dynamic light scattering (DLS) spectra quantitively display the size distributions of PbI2 colloids in the precursor solutions with variant H2O concentrations, which correspond well to the conductance results. The mean size of the colloids is significantly reduced to 172.8 nm with H2O addition (x  0.02) in comparison with 432.9 nm (x  0) and 978.3 nm (control). We note that the x  0.04 precursor with the probably higher proton activity unexpectedly shows an increased average size of 492.0 nm even larger than the x  0 sample, which may result from the decreased solubility of PbI2 upon excessive H2O addition. The colloid size in the precursor solutions correlate closely with the crystallite size of the final perovskite films (Figure 2f). All the above results well explain the effect of H2O molecules on the morphology evolution. In light of the impact of the crystallite size on the crystallographic phase and structural stability, we assess the size-dependent phase stability of the CsPbI3 polymorphs by first-principles density functional theory (DFT) calculations. The thermodynamic stability of a particle system is dictated by the Gibbs free energy (G)

comprising the bulk free energy (Gb) and the surface free energy (Gs): G  Gb  Gs. Given that the portion of atoms on the surface becomes much higher as the crystallite dimensions are diminished, inequality in the surface energy can favor the formation of a particular polymorph.44-47 The calculated free energies of the lowMiller-index surfaces of the perovskite-structured -CsPbI3 are markedly lower than those of the nonperovskite CsPbI3 (Table S4 and Figure S5). Since the distributions of surface planes exposed of the crystallites with an irregular geometric shape cannot be quantified, mathematical expectations are employed to evaluate the weighted surface energy based on the assumption that the crystal faces are exposed in a random manner. We plot the calculated Gibbs free energy relative to bulk -CsPbI3 as a function of surface area in Figure 2g where the slope of each line gives the surface free energy meanwhile the difference at zero surface area represents the difference of their bulk free energy of 20.9 kJ/mol. The disparity in the surface energy between -CsPbI3 (0.13 J/m2) and -CsPbI3 (2.57 J/m2) leads to a reversal in the relative magnitudes of their Gibbs free energies at surface areas greater than ~8,600 m2/mol, which corresponds to an average crystallite size of ~100 nm (the geometry of a crystallite approximates to a cube), and therefore changes their relative phase stabilities. The theoretical outputs are highly consistent with our experimental observations that the -CsPbI3 crystals are stabilized at remarkably reduced dimensions of ~100 nm. Our study demonstrates that the polymorph becomes thermodynamically -CsPbI3 preferred over -CsPbI3 at significantly extended surface areas when the crystallite is trimmed down to nanoscale. It is important to note that the -CsPbI3 thin films are of intrinsic thermodynamic stability rather than the phase sustainability derived from the long-chain organic ligands or the modified perovskite compositions which may create uncertainty to the thermal reliability and the electronic structure suitability. On the basis of the above results and discussion, we propose the plausible mechanism for the formation of the thermodynamically stable -CsPbI3 thin films through the solution process, as illustrated in Figure 3. Initially the

5



Page 6 of 12

Figure 4. Optoelectronic properties of -phase CsPbI3. (a) Total and partial density of states (DOS) for -phase CsPbI3. Charge density associated with (b) the valence band maximum (VBM) and (c) the conduction band minimum (CBM) of -CsPbI3. (d) Electronic band structure of -CsPbI3 calculated at the HSE+SOC level of theory (colored) and within the GGA only (grey) along high-symmetry directions in the Brillouin zone. (e) Photoluminescence (PL) spectra of CsPbI3 films prepared with different H2O concentrations (x). (f) J‐V curves of photovoltaic devices based on the corresponding perovskite films under AM 1.5 illumination (100 mW cm-2).

precursor solution is composed of PbI2 colloids and ions solvated in the solvent mixture DMF/HI with the presence of the molecular complex DMFxHI. At the second stage, the colloids are split up due to the increased activity of H and I released from the H2O-induced proton transfer process. During spin coating, I in the solution coordinate to Pb2 of the reduced colloids to form the corner-sharing [PbI6]4 octahedra intercalated by Cs. Finally, the octahedral framework is converted to the perovskite-structured polymorph with decreased crystallite size after annealing at 100 ℃ , stabilizing the orthorhombic -CsPbI3 polycrystalline thin films with surface areas considerably enlarged.

further reveal that the VBM is an antibonding state from the hybridization between the Pb 6s orbitals and the I 5p orbitals while the CBM is almost a nonbonding state dominated by the Pb 6p orbitals. The minor octahedral tilting in the lattice of -CsPbI3, with the PbIPb bond angle (155.6°) much larger than those in -CsPbI3 (95.09° and 91.40°),49 gives rise to the strong antibonding state from the appreciable Pb-I overlap. It is found that this character is responsible for keeping the defect tolerance of -CsPbI3, which is critical for the high performance of PSCs.49 On the other hand, the strong antibonding interaction in -CsPbI3 destabilizes the VBM, elevating its energy. The CBM, by contrast, is supposed to respond less intensely to structural changes than the VBM due to the nonbonding character. As a result, the orthorhombic CsPbI3 is expected to possess a substantially narrower bandgap than the orthorhombic  phase though slightly broader than the cubic  phase. The DFT band structure including spin-orbit coupling (SOC) of the  phase in Figure 4d corroborates this prediction by determining a direct bandgap of 1.96 eV at the  point in comparison with the estimated bandgap of 2.34 eV for -CsPbI3 and 1.63 eV for -CsPbI3 (Figure S6 and Table S5). The calculations were performed using the Heyd-Scuseria-Ernzerhof (HSE) hybrid functional50 including SOC, as implemented in the PWmat code51-52, in order to improve the accuracy of the

Theoretical and Experimental Optoelectronics of  Phase. To ascertain the eligibility of the black -CsPbI3 for photovoltaic applications, we performed ab initio DFT calculations using the CASTEP code48 to investigate its electronic structure. In Figure 4a we show the total density of states (DOS) and partial density of states projected on Cs, Pb and I atoms, respectively. The DOS clearly display that the band edge states predominantly arises from the I p orbitals and the Pb p orbitals. In sharp contrast, Cs scarcely contributes to the states near the Fermi level, suggesting the strong ionic nature of the bonding with Cs. The charge density originating from the valence band maximum (VBM) and the conduction band minimum (CBM) in Figure 4b,c

6


Page 7 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Photovoltaic performance characteristics. (a) J‐V curve of the best-performing -CsPbI3 solar cell (x = 0.02) under simulated AM 1.5 sunlight of 100 mW cm-2 irradiation. (b) Histograms of the power conversion efficiency (PCE) values (based on 20 devices each), (c) external quantum efficiency (EQE) spectra along with the integrated products of the EQE curves and (d) stability plots of PCE as a function of storage time for -CsPbI3-based devices (x = 0.02) and reference devices (x = 0). The device stability was measured in ambient environment without encapsulation. (e) Stabilized power output (SPO) of the champion CsPbI3 device. The SPO was measured in ambient air at a constant forward bias of 0.78 V under continuous 1-sun illumination.

bandgap which is usually underestimated within the generalized gradient approximation (GGA). The photoluminescence (PL) spectrum of the -phase film (x  0.02) (Figure 4e) exhibits a band edge emission at 709 nm, corresponding to a bandgap of 1.75 eV. In brief, semiconducting -CsPbI3 of such electronic structure shows good potential for an efficient solar cell, given the defect-tolerant properties and the modest difference with the optimal bandgap of ∼1.3 eV that yields the maximum PCE of the Shockley-Queisser limit for single-junction devices.53-54

caused by grain boundaries is relatively benign to carrier transport in CH3NH3PbI3 films.56 Besides, a blueshift can be discerned along with the increase of the PL intensity as the crystallite size decreases, which might be ascribed to the higher fraction of atoms on the surface free from spin-orbit coupling (SOC) that narrows the bandgap.57 The planar heterojunction architecture was then deployed to prove the optoelectronic functionality of the -phase thin films in solar-cell devices. We can see from the current densityvoltage (J‐V) curves in Figure 4f that the -CsPbI3 film (x  0.02) demonstrates superior photovoltaic response to the solar radiation with significantly enhanced open-circuit voltage (Voc) and short-circuit current density (Jsc) as compared with other samples prepared with varying H2O content. It is worth emphasizing that the orthorhombic CsPbI3 exhibits comparable optoelectronic performance with the cubic -CsPbI3 that has been extensively reported.12-13, 21, 27

Besides, strong radiative recombination from the CsPbI3 film (x  0.02) with the PL intensity an order of magnitude greater than other samples (Figure 4e) indicates that surface dangling bonds do not result in notable midgap trap states. This observation is in good agreement with the exceptionally high quantum yields and carrier mobility of CsPbX3 nanocrystals,20, 55 and gives a positive response to Chu et al. who show that surface

Photovoltaic Performance of ‐CsPbI3 Devices. We

7



for a planar heterojunction device.39 Furthermore, the CsPbI3-based device maintains nearly 100% of its initial performance after continuous operation at a constant forward bias of 0.78 V (MPP) under 1-sun illumination for more than 7000 seconds, showing substantially improved stability over the CH3NH3PbI3-based device (Figure S11). The excellent operational stability may well relate to the all-inorganic structure of -CsPbI3 contrasting with the CH3NH3 moiety of smaller mass and higher rotational and vibrational degrees of freedom that could account for the mounting phonon scattering under continuous solar irradiation (to which a rising temperature is incidental) and hence decrease the carrier mobility.55

Table 1. Statistical photovoltaic parameters of CsPbI3 solar cells with different H2O concentrations (x). x

Voc (V)

Jsc (mA cm‐2)

FF (%)

PCE (%)

0

0.79  0.09

12.46  1.55

29.5  4.8

2.91  0.63

0.01

0.97  0.05

13.96  0.48

59.3  2.8

7.99  0.38

0.02

1.01  0.04

15.12  0.92

60.1  2.7 10.15  0.77

0.04

0.97  0.04

13.51  0.47

47.6  6.1

Page 8 of 12

6.23  0.89

further confirmed the photovoltaic performance of the CsPbI3 PSCs fabricated on a larger scale (12 to 30 cells for each type). As shown in Table 1, all the photovoltaic parameters, namely Voc, Jsc, fill factors (FF) and PCE, are markedly improved in the -CsPbI3-based devices (x  0.02). In particular, the average FF is approximately doubled from 29.5% (x  0) to 60.1% (x  0.02). The overall enhancement in performance for the planar devices can be attributed to the high purity of the black  phase as well as the improved film morphology of reduced pinholes and roughness (Figure 2a-c). We show in Figure 5a the J‐V scan of the champion device fabricated on the -CsPbI3 thin film measured under simulated AM 1.5 sunlight (100 mW cm-2) in ambient air, displaying outstanding performance with Voc  1.04 V, Jsc  16.53 mA cm-2, FF  65.7% and PCE  11.30%. To the best of our knowledge, this is one of the highest PCEs for CsPbI3 thin-film solar cells by far and, of course, the first report of the -CsPbI3-based PSCs.

■ CONCLUSION In summary, we report a new perovskite-structured thin film of all-inorganic perovskites, the orthorhombic CsPbI3 thin film, prepared via facile one-step solution coating for highly efficient and stable solar cells. By introducing a small amount of H2O, the proton transfer process in the CsPbI3 precursor solution is triggered and then dictates the size-driven phase formation of CsPbI3 perovskite. We theoretically and experimentally demonstrate that -CsPbI3 has a significantly lower surface energy than -CsPbI3, thus thermodynamically favored as the surface area is enlarged above 8,600 m2/mol. More importantly, the desirable electronic structure and optoelectronic performance of -CsPbI3 thin films contribute to a PCE of 11.3%, among the best reported PCEs for CsPbI3 thin-film solar cells to date. By virtue of the improved phase stability, the -CsPbI3-based device exhibits substantially enhanced reliability without noticeable loss in performance under ambient environment for months and continuous operating conditions for hours. Our study offers a novel and fundamental perspective to address the pressing concern, the structural stability, of black-phase CsPbI3 perovskite. Such strategy could also be employed to stabilize the target phase of other polycrystalline materials. Moreover, the thermodynamically stable -CsPbI3 thin films presented herein open up new possibilities for high-performance photovoltaic and other optoelectronic devices based on stabilized all-inorganic perovskites.

We compared the optimized devices of -CsPbI3 perovskite with those based on the pristine films (x  0) as a reference in Figure 5b-d. The device performance was highly reproducible, showing a striking improvement of the average PCE by more than threefold (Figure 5b and Figure S7). The good reproducibility can be credited to the simple film fabrication method of one-step spin coating and low-temperature annealing. Moreover, the -CsPbI3 solar cell exhibits an increase in external quantum efficiency (EQE) (Figure 5c) over a wide spectrum. The current densities integrated from the EQE spectra for the typical -CsPbI3 and reference device are consistent with the Jsc extracted from the corresponding J‐V curves (Figure S8). The device stability was tested without any encapsulation in ambient conditions (temperature: 25  5℃, relative humidity: 10  5%), as shown in Figure 5d. Owing to the thermodynamically stable thin films of the  phase, the corresponding device demonstrates remarkably enhanced stability compared to the reference device over the course of more than 40 days, without any performance loss relative to its initial value but with a modicum of improvement as observed in the -CsPbI3 QD solar cells 21 (Figure S9). We also found hysteresis between forward and backward scan in our planar solar cells (Figure S10). It is noted that the stabilized power output (SPO) measurement at the maximum power point (MPP) is a better approach to evaluate the hysteresis behavior.39 In Figure 5e, the MPP tracking of the best-performing CsPbI3 solar cell indicates a PCE of ~9.7%, 86% of the best J‐V scan efficiency, of which the hysteresis is relatively low

■ EXPERIMENTAL SECTION Precursor Preparation. The CsPbI3 precursor solution was prepared by dissolving 0.48 mmol of lead iodide (PbI2, 99.999%) and 0.48 mmol of cesium iodide (CsI, 99.999%) in the mixed solvent of 1 ml of anhydrous N,Ndimethylformamide (DMF, 99.8%) and 33 L of hydroiodic acid (HI, 57 wt% in H2O). Prior to spin coating, different amounts (x) of deionized H2O (x: volume ratio versus DMF) are added to the precursor solution. All materials were purchased from Sigma-Aldrich and used as received without any purification. Perovskite Deposition. The perovskite films were prepared by spin-coating the CsPbI3 precursor solution at 5,000 r.p.m. for 30 s, followed by thermal annealing at 100℃ for 5 min. For the control sample, the solvent of the

8


Page 9 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


𝛥𝐸 𝛥𝐴 where E represents the free energy difference between the systems before and after new surfaces are generated, and A is the area of the new surfaces. For a slab, the surface energy can be determined as:58-60

precursor solution was only DMF without HI and films were annealed at 335℃ for 5 min.

𝛾

Device Fabrication. The FTO glass was treated by UVozone cleaner for 15 min, followed by spay pyrolysis of 20 mM titanium diisopropoxide bis(acetylacetonate) in ethanol at 450℃ and sintered at 500℃ for 30 min. The resulting TiO2 layer was then immersed in 20mM TiCl4 aqueous solution at 70℃ for 30 min, followed by sintering at 500℃ for 30 min to produce a compact TiO2 layer on the FTO glass. The perovskite film was then deposited on the FTO/compact-TiO2 substrate as mentioned above. Poly(3hexylthiopehne) (P3HT) chlorobenzene solution (12 mg ml-1) was spin-coated on the perovskite layer at 3,000 r.p.m. for 30 s. Finally, an 80-nm gold electrode was thermally evaporated on the P3HT layer to complete the device. Characterizations. X-ray diffraction (XRD) spectra were collected at room temperature from samples of perovskite deposited on the FTO glass by a PANalytical Empyrean Powder X-ray diffractometer using Cu K1 radiation (1.541 Å) operating at 40 kV and 40 mA. The fresh samples were placed in a sealed chamber in N2 atmosphere during the measurement. Rietveld refinement of the XRD data was performed using FullProf software. The stability test of the perovskite films was conducted by storing the samples in ambient conditions (temperature: 25  5℃, relative humidity: 10  5%). Scanning electron microscopy (SEM) images were obtained using Hitachi SU-8020. The average crystallite size was estimated from SEM images using Nano Measure software. Atomic Force Microscopy (AFM) images were obtained by Bruker Demension Icon. The surface roughness was measured as the root meansquared roughness over the scanning area. The electrical conductivity measurement was performed using JENWAY Model 4320 conductivity meter at 20℃. The dynamic light scattering (DLS) measurement was performed using a zeta potential analyzer (Zetasizer Nano ZS, Malvern). The crystallite size was estimated by SEM images using Nano Measure software. Steady-state photoluminescence (PL) spectra were obtained by Edinburgh Instrument FLS 980 with an excitation wavelength of 510 nm. Ultravioletvisible (UV-vis) absorption spectra were collected by an UV-vis spectrophotometer (UH4150, HITACHI). A solar simulator (450W Model 91150, Newport) with AM 1.5 sunlight was used as the irradiation source (100 mW cm-2, calibrated with a NREL reference cell). The current density-voltage (J‐V) curves and the device stability were measured using Keithley 2420 in ambient atmosphere (temperature: 25  5℃, relative humidity: 10  5%) without encapsulation. The active areas of the device were defined by the cross area of top and bottom electrodes and a black metal mask with an area of 0.09 cm2. The mask areas were measured using a microscope (ECLIPSE LV150N, Nicon). The external quantum efficiency (EQE) measurement was performed using a QE-R3011 measurement system (Enli Technology). Surface Free Energy Calculations. We used the following equation to calculate the surface free energy  of a (hkl) crystal face:

𝐸

𝑁𝐸 2𝐴 where N is the number of layers into which the slab is is the formation energy of the N-layer slab, divided, 𝐸 represents the formation energy of an identical slab 𝐸 but undivided, which can be treated as a monolayer, and A is the area of the generated surfaces. Our calculation was based on the density functional theory (DFT) using the PWmat code with the pseudopotentials that adopt the USPP-GBRV.51-52 In all our models, the width of the vacuum region was set as 8 Å, along with the N value of 5 and K point grids of 2  2  2. Experimentally determined lattice constants and atomic coordinates were employed. Electronic Structure Calculations. The electronic structure of -CsPbI3 was calculated by density functional theory (DFT) using the CASTEP program code48 with the plane-wave pseudopotential method. We adopted the generalized gradient approximation (GGA) in the form of the Perdew-Burke-Ernzerhof (PBE) for the exchangecorrelation potentials.61 The ultrasoft scalar relativistic pseudopotential was used with a plane-wave energy cutoff of 300 eV. The first Brillouin zone was sampled with grid spacing of 0.037 Å-1.62 The self-consistent field was used with a tolerance of 5  10-7 eV/atom. To improve the accuracy of the bandgap which is usually underestimated within the GGA, the Heyd-Scuseria-Ernzerhof (HSE) hybrid functional50 was adopted including spin-orbit coupling (SOC) using the PWmat code51-52 with the NCPP-SG15-PBESOC63-65 pseudopotential. The lattice constants and atomic coordinates used in the calculations were derived from the experimental results. 𝛾

ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge on the ACS Publications website. Film appearance, spectral and structural characterizations, surface models, electronic band structure calculations, statistical photovoltaic parameters, J-V scans, phase and device stability test, review of inorganic perovskite solar cells, refined crystallographic data, indexing details, surface energy calculations, bandgap comparison.

AUTHOR INFORMATION Corresponding Author *[email protected] *[email protected]

Notes The authors declare no competing financial interests.

ACKNOWLEDGMENTS

9



Page 10 of 12

(24) Wang, Q.; Jin, Z.; Chen, D.; Bai, D.; Bian, H.; Sun, J.; Zhu, G.; Wang, G.; Liu, S. F. Adv. Energy Mater. 2018, 8, 1800007. (25) Eperon, G. E.; Paterno, G. M.; Sutton, R. J.; Zampetti, A.; Haghighirad, A. A.; Cacialli, F.; Snaith, H. J. J. Mater. Chem. A 2015, 3, 19688-19695. (26) Luo, P.; Xia, W.; Zhou, S.; Sun, L.; Cheng, J.; Xu, C.; Lu, Y. J. Phys. Chem. Lett. 2016, 7, 3603-3608. (27) Zhang, T.; Dar, M. I.; Li, G.; Xu, F.; Guo, N.; Grä tzel, M.; Zhao, Y. Sci. Adv. 2017, 3, e1700841. (28) Goldschmidt, V. M. Naturwissenschaften 1926, 14, 447-485. (29) Saparov, B.; Mitzi, D. B. Chem. Rev. 2016, 116, 4558-4596. (30) Lee, Y.; Mitzi, D. B.; Barnes, P. W.; Vogt, T. Phys. Rev. B 2003, 68, 020103. (31) Trots, D. M.; Myagkota, S. V. J. Phys. Chem. Solids 2008, 69, 2520-2526. (32) Marronnier, A.; Roma, G.; Boyer-Richard, S.; Pedesseau, L.; Jancu, J. M.; Bonnassieux, Y.; Katan, C.; Stoumpos, C. C.; Kanatzidis, M. G.; Even, J. ACS Nano 2018, 12, 3477-3486. (33) Sutton, R. J.; Filip, M. R.; Haghighirad, A. A.; Sakai, N.; Wenger, B.; Giustino, F.; Snaith, H. J. ACS Energy Lett. 2018, 1787-1794. (34) Ohta, H.; Harada, J.; Hirotsu, S. Solid State Commun. 1973, 13, 1969-1972. (35) Hirotsu, S.; Harada, J.; Iizumi, M.; Gesi, K. J. Phys. Soc. Jpn. 1974, 37, 1393-1398. (36) Chung, I.; Song, J. H.; Im, J.; Androulakis, J.; Malliakas, C. D.; Li, H.; Freeman, A. J.; Kenney, J. T.; Kanatzidis, M. G. J. Am. Chem. Soc. 2012, 134, 8579-8587. (37) Onoda-Yamamuro, N.; Yamamuro, O.; Matsuo, T.; Suga, H. J. Phys. Chem. Solids 1992, 53, 277-281. (38) Fabini, D. H.; Stoumpos, C. C.; Laurita, G.; Kaltzoglou, A.; Kontos, A. G.; Falaras, P.; Kanatzidis, M. G.; Seshadri, R. Angew. Chem. Int. Ed. 2016, 55, 15392-15396. (39) Zhang, W.; Saliba, M.; Moore, D. T.; Pathak, S. K.; Horantner, M. T.; Stergiopoulos, T.; Stranks, S. D.; Eperon, G. E.; AlexanderWebber, J. A.; Abate, A.; Sadhanala, A.; Yao, S.; Chen, Y.; Friend, R. H.; Estroff, L. A.; Wiesner, U.; Snaith, H. J. Nat. Commun. 2015, 6, 6142. (40) Yan, K.; Long, M.; Zhang, T.; Wei, Z.; Chen, H.; Yang, S.; Xu, J. J. Am. Chem. Soc. 2015, 137, 4460-4468. (41) Nayak, P. K.; Moore, D. T.; Wenger, B.; Nayak, S.; Haghighirad, A. A.; Fineberg, A.; Noel, N. K.; Reid, O. G.; Rumbles, G.; Kukura, P.; Vincent, K. A.; Snaith, H. J. Nat. Commun. 2016, 7, 13303. (42) Sears, P. G.; Wolford, R. K.; Dawson, L. R. J. Electrochem. Soc. 1956, 103, 633-636. (43) Thomas, A. B.; Rochow, E. G. J. Am. Chem. Soc. 1957, 79, 1843-1848. (44) McHale, J. M.; Auroux, A.; Perrotta, A. J.; Navrotsky, A. Science 1997, 277, 788-791. (45) Fu, D.; Suzuki, H.; Ishikawa, K. Phys. Rev. B 2000, 62, 31253129. (46) Wang, F.; Han, Y.; Lim, C. S.; Lu, Y.; Wang, J.; Xu, J.; Chen, H.; Zhang, C.; Hong, M.; Liu, X. Nature 2010, 463, 1061-1065. (47) Fu, Y.; Wu, T.; Wang, J.; Zhai, J.; Shearer, M. J.; Zhao, Y.; Hamers, R. J.; Kan, E.; Deng, K.; Zhu, X. Y.; Jin, S. Nano Lett. 2017, 17, 4405-4414. (48) Clark, S. J.; Segall, M. D.; Pickard, C. J.; Hasnip, P. J.; Probert, M. J.; Refson, K.; Payne, M. C. Z. Kristallogr. 2005, 220, 567-570. (49) Huang, Y.; Yin, W.-J.; He, Y. J. Phys. Chem. C 2018, 122, 13451350. (50) Heyd, J.; Scuseria, G. E.; Ernzerhof, M. J. Chem. Phys. 2006, 124, 219906. (51) Jia, W.; Fu, J.; Cao, Z.; Wang, L.; Chi, X.; Gao, W.; Wang, L. W. J. Comput. Phys. 2013, 251, 102-115. (52) Jia, W.; Cao, Z.; Wang, L.; Fu, J.; Chi, X.; Gao, W.; Wang, L. W. Comput. Phys. Comm. 2013, 184, 9-18. (53) Shockley, W.; Queisser, H. J. J. Appl. Phys. 1961, 32, 510-519. (54) Wadia, C.; Alivisatos, A. P.; Kammen, D. M. Environ. Sci. Technol. 2009, 43, 2072–2077. (55) Yettapu, G. R.; Talukdar, D.; Sarkar, S.; Swarnkar, A.; Nag, A.; Ghosh, P.; Mandal, P. Nano Lett. 2016, 16, 4838-4848.

We acknowledge the financial support from the National Natural Science Foundation of China (21573249), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB12020100) and the Youth Innovation Promotion Association CAS (2017050). We thank Yang Sun at ICCAS for assistance with XRD measurements and Jun Deng for assistance with DFT calculations.

REFERENCES (1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am. Chem. Soc. 2009, 131, 6050-6051. (2) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Science 2012, 338, 643-647. (3) Burschka, J.; Pellet, N.; Moon, S. J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Grä tzel, M. Nature 2013, 499, 316-319. (4) Zhou, H.; Chen, Q.; Li, G.; Luo, S.; Song, T. B.; Duan, H. S.; Hong, Z.; You, J.; Liu, Y.; Yang, Y. Science 2014, 345, 542-546. (5) Tan, H.; Jain, A.; Voznyy, O.; Lan, X.; Garcia de Arquer, F. P.; Fan, J. Z.; Quintero-Bermudez, R.; Yuan, M.; Zhang, B.; Zhao, Y.; Fan, F.; Li, P.; Quan, L. N.; Zhao, Y.; Lu, Z. H.; Yang, Z.; Hoogland, S.; Sargent, E. H. Science 2017, 355, 722-726. (6) Yang, W. S.; Park, B. W.; Jung, E. H.; Jeon, N. J.; Kim, Y. C.; Lee, D. U.; Shin, S. S.; Seo, J.; Kim, E. K.; Noh, J. H.; Seok, S. I. Science 2017, 356, 1376-1379. (7) Nagabhushana, G. P.; Shivaramaiah, R.; Navrotsky, A. Proc. Natl. Acad. Sci. 2016, 113, 7717-7721. (8) Conings, B.; Drijkoningen, J.; Gauquelin, N.; Babayigit, A.; D'Haen, J.; D'Olieslaeger, L.; Ethirajan, A.; Verbeeck, J.; Manca, J.; Mosconi, E.; Angelis, F. D.; Boyen, H.-G. Adv. Energy Mater. 2015, 5, 1500477. (9) Sutton, R. J.; Eperon, G. E.; Miranda, L.; Parrott, E. S.; Kamino, B. A.; Patel, J. B.; Hörantner, M. T.; Johnston, M. B.; Haghighirad, A. A.; Moore, D. T.; Snaith, H. J. Adv. Energy Mater. 2016, 6, 1502458. (10) Stoumpos, C. C.; Malliakas, C. D.; Kanatzidis, M. G. Inorg. Chem. 2013, 52, 9019-9038. (11) Sharma, S.; Weiden, N.; Weiss, A. Z. Phys. Chem. 1992, 175, 63-80. (12) Hu, Y.; Bai, F.; Liu, X.; Ji, Q.; Miao, X.; Qiu, T.; Zhang, S. ACS Energy Lett. 2017, 2, 2219-2227. (13) Wang, Q.; Zheng, X.; Deng, Y.; Zhao, J.; Chen, Z.; Huang, J. Joule 2017, 1, 371-382. (14) Beal, R. E.; Slotcavage, D. J.; Leijtens, T.; Bowring, A. R.; Belisle, R. A.; Nguyen, W. H.; Burkhard, G. F.; Hoke, E. T.; McGehee, M. D. J. Phys. Chem. Lett. 2016, 7, 746-751. (15) Zhang, J.; Bai, D.; Jin, Z.; Bian, H.; Wang, K.; Sun, J.; Wang, Q.; Liu, S. F. Adv. Energy Mater. 2018, 8, 1703246. (16) Zeng, Q.; Zhang, X.; Feng, X.; Lu, S.; Chen, Z.; Yong, X.; Redfern, S. A. T.; Wei, H.; Wang, H.; Shen, H.; Zhang, W.; Zheng, W.; Zhang, H.; Tse, J. S.; Yang, B. Adv. Mater. 2018, 30. (17) Liu, C.; Li, W.; Zhang, C.; Ma, Y.; Fan, J.; Mai, Y. J. Am. Chem. Soc. 2018, 140, 3825-3828. (18) Bian, H.; Bai, D.; Jin, Z.; Wang, K.; Liang, L.; Wang, H.; Zhang, J.; Wang, Q.; Liu, S. F. Joule 2018, 2, 1-11. (19) Shen, Q.; Ripolles, T. S.; Even, J.; Zhang, Y.; Ding, C.; Liu, F.; Izuishi, T.; Nakazawa, N.; Toyoda, T.; Ogomi, Y.; Hayase, S. J. Energy Chem. 2018, 27, 1170-1174. (20) Protesescu, L.; Yakunin, S.; Bodnarchuk, M. I.; Krieg, F.; Caputo, R.; Hendon, C. H.; Yang, R. X.; Walsh, A.; Kovalenko, M. V. Nano Lett. 2015, 15, 3692-3696. (21) Swarnkar, A.; Marshall, A. R.; Sanehira, E. M.; Chernomordik, B. D.; Moore, D. T.; Christians, J. A.; Chakrabarti, T.; Luther, J. M. Science 2016, 354, 92-95. (22) Fu, Y.; Rea, M. T.; Chen, J.; Morrow, D. J.; Hautzinger, M. P.; Zhao, Y.; Pan, D.; Manger, L. H.; Wright, J. C.; Goldsmith, R. H.; Jin, S. Chem. Mater. 2017, 29, 8385-8394. (23) Sanehira, E. M.; Marshall, A. R.; Christians, J. A.; Harvey, S. P.; Ciesielski, P. N.; Wheeler, L. M.; Schulz, P.; Lin, L. Y.; Beard, M. C.; Luther, J. M. Sci. Adv. 2017, 3, eaao4204.

10


Page 11 of 12 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(56) Chu, Z.; Yang, M.; Schulz, P.; Wu, D.; Ma, X.; Seifert, E.; Sun, L.; Li, X.; Zhu, K.; Lai, K. Nat. Commun. 2017, 8, 2230. (57) Brandt, R. E.; Stevanović, V.; Ginley, D. S.; Buonassisi, T. MRS Commun. 2015, 5, 265-275. (58) Boettger, J. C. Phys. Rev. B 1994, 49, 16798-16800. (59) Fiorentini, V.; Methfessel, M. J. Phys.: Condens. Matter 1996, 8, 6525-6529. (60) Bruno, M.; Prencipe, M. Cryst. Eng. Comm. 2013, 15, 67366744. (61) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865-3868. (62) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188-5192. (63) Hamann, D. R. Phys. Rev. B 2013, 88, 085117. (64) Schlipf, M.; Gygi, F. Comput. Phys. Comm. 2015, 196, 36-44. (65) Scherpelz, P.; Govoni, M.; Hamada, I.; Galli, G. J. Chem. Theory Comput. 2016, 12, 3523–3544.

11



Page 12 of 12

For Table of Contents Only

12


Thermodynamically Stable Orthorhombic Î³-CsPbI3 Thin Films for High

Recommend Documents