Self-Alignment of the Methylammonium Cations in Thin-Film

Aug 4, 2014 - Devices and Collaborative Innovation Center of Suzhou Nano Science and ... Radiation Research, 199 Ren'ai Road, Suzhou Industrial Park,...
0 downloads 0 Views 974KB Size
Download PDF
Letter pubs.acs.org/JPCL

Self-Alignment of the Methylammonium Cations in Thin-Film Organometal Perovskites John A. McLeod,*,†,‡ Zhongwei Wu,† Pengfei Shen,†,‡ Baoquan Sun,† and Lijia Liu*,†,‡ †

Institute of Functional Nano and Soft Materials (FUNSOM), Jiangsu Key Laboratory for Carbon-Based Functional Materials and Devices and Collaborative Innovation Center of Suzhou Nano Science and Technology, Soochow University, 199 Ren'ai Road, Suzhou Industrial Park, Suzhou, Jiangsu 215123, China ‡ Soochow University-Western University Center for Synchrotron Radiation Research, 199 Ren'ai Road, Suzhou Industrial Park, Suzhou, Jiangsu 215123, China S Supporting Information *

ABSTRACT: A comparative study of the electronic structure of methylammonium (CH3NH3) in organometallic lead triiodide perovskite (CH3NH3PbI3) thin films synthesized using either one- or two-step deposition protocols is performed using angleresolved C K-edge soft X-ray absorption spectroscopy (XAS) and model calculations. We find that our XAS measurements can be accurately related to the ground-state unoccupied orbitals using a simple crystal field model. We further find that films made by the one-step deposition protocol exhibit angle-dependent features, indicating long-range alignment of the CH3NH3 molecules, although the angle-dependency decreases as the film thickness increases. No angle-dependency was observed in the films made via the two-step deposition method.

SECTION: Energy Conversion and Storage; Energy and Charge Transport

O

corner-sharing BX6 octahedra centered on the crystal cell vertices, while the A is located at the center of the cubic lattice. However, because of the large atomic size of the iodine atoms, the PbI6 octahedra in CH3NH3PbI3 are distorted and tilted, and the symmetry is lowered to a tetragonal phase. Most of the fundamental research thus far has been focused on the octahedral component of the perovskites, such as tuning the structure and properties by mixing halides or by metal substitution, while relatively little attention has been paid to the organic cations. Indeed, to simplify structural analysis, many studies treated the organic cations as hard spheres that resemble the inorganic CsSnI3 perovskite and consider the cations as free isotropic rotators within the perovskite framework.10,11 It is nevertheless worthwhile to investigate the local structure of the CH3NH3 to determine if they are indeed random or if they exhibit any long-range orientation. This line of research is motivated by several reasons, one being that the structure of these perovskites has been found to be highly temperature-dependent. It has been reported that as the temperature decreases methylammonium perovskite undergoes a cubic−tetragonal−orthorhombic phase transition in which the CH3NH3 become increasingly confined until the only rotation is around the C−N axis.12,13 If the CH3NH3 was highly

rganometallic perovskites are inorganic−organic hybrids containing divalent metal halides anions and alkylammonium cations. Studies on the structure and properties of such perovskite compounds date back to the 1990s,1−4 when the magnetic, optical, and electric properties of organic− inorganic perovskites were comparatively investigated with respect to their inorganic counterparts. However, it was not until 2009, when methylammonium lead triiodide perovskite (CH3NH3PbI3) was implemented as the light sensitizer in photovoltaic cells, that this class of material rapidly attracted tremendous attention in the solar energy research community.5 More recently, it was demonstrated that a photovoltaic device incorporating CH3NH3PbI3 combined with a solid hole transport material exhibited enhanced stability and a power conversion efficiency of 9.7%.6 Since then, many studies have been focusing on improving device performance and efficiency by optimizing the structure (this includes layer configurations as well as fabrication techniques) of solar cells based on archetypal CH3NH3PbI3 as well as related compounds (such as mixed halides CH3NH3PbI3−xClx). These studies have paid off, and device efficiency has rapidly increased to 15% and higher.7−9 It is not only essential to optimize the fabrication procedure but also crucial to understand the fundamental properties of the perovskite material itself to improve the device performance of cells employing these perovskites. It is well known that an ideal ABX3 perovskite compound has a cubic structure, consisting of © 2014 American Chemical Society

Received: July 15, 2014 Accepted: August 4, 2014 Published: August 4, 2014 2863

dx.doi.org/10.1021/jz501472d | J. Phys. Chem. Lett. 2014, 5, 2863−2867

The Journal of Physical Chemistry Letters

Letter

disordered at ambient temperatures, this would frustrate the phase transition. Second, it has been pointed out that hydrogen bonding between the NH3 head and the PbI6 is not negligible and could potentially result in the formation of long-range order.1,14,15 We study the orientation of the CH3NH3 near the surface of three different preparations of archetypal perovskite CH3NH3PbI3 thin films. These thin films are synthesized using two- (sample #1) and one-step (samples #2 and #3) solution-deposition methods, with thicknesses of 280, 400, and 260 nm for samples #1, #2, and #3, respectively. Both methods utilize the same precursors (CH3NH3I and PbI2) and are reported to be effective ways of producing high-quality films.6,16,17 We examine the electronic structure and possible alignment of CH3NH3 by angle-dependent X-ray absorption spectroscopy (XAS) measurements at the C K-edge (excitation of C 1s core electrons), and we present evidence that the C−N bond axis of the CH3NH3 is parallel to the surface of films synthesized by one-step deposition, indicating some degree of long-range order. Detailed information on sample preparation and characterization can be found in the Experimental Section and the Supporting Information. The XRD pattern from the three films is shown in Figure 1. All films exhibit the characteristic features of the tetragonal

Figure 2. Measured XAS spectra of the perovskite films from the C 1s edge. The “normal” and “grazing” spectra were acquired with the X-ray beam at an incidence angle of 0 and 70° with respect to the sample normal, respectively.

do not. In particular, the spectral intensity of the LUMO in samples #2 and #3 decreases rapidly as the X-ray beam moves toward a grazing incidence angle (see the feature near 285 eV in Figure 2), while an additional feature near 288 eV appears at grazing incidence. (This feature is only prominent in sample #3; it is merely a shoulder in sample #2.) The evolution of relative peak intensity at a series of incident angles from normal to gracing incidence can be found in Figure S3 in the Supporting Information. The angle dependence of the C 1s XAS for samples #2 and #3 suggests that CH3NH3 are aligned (at least near the surface; the probe depth of TEY XAS is only a few nanometers21). The influence of CH3NH3 alignment on the X-ray spectra is illustrated schematically in Figure 3. In this schematic, the

Figure 1. XRD patterns of the three perovskite thin films. The features marked with asterisks are the (001) and (101) peaks from PbI2 of a hematite structure. XRD of CH3NH3I and PEDOT:PSS−PMMAcoated ITO are also included as reference. Figure 3. Schematic of the orientation of CH3NH3 relative to the incident X-ray beam, illustrating the surface normal n, the C−N bond axis z, the incidence angle of the X-ray beam, φ, and the azimuthal angle, θ, of the X-ray beam relative to z. The X-ray beam is the green line; the electric field of the X-ray beam is the wavy green line. (a) Situation at normal incidence. (b) Situation at grazing incidence.

perovskite structure. The three major diffraction peaks at 2θ values of 14.09, 28.39, and 31.90° correspond to the (110), (220), and (310) crystal planes, respectively. There is some unreacted PbI2 present in sample #1 (unsurprising, given the two-step synthesis method), indicated by its characteristic (001) diffraction peak at 12.8°.18,19 However, importantly, we observe no unreacted CH3NH3I in any of the films, indicating that the reaction or decomposition of the CH3NH3I during annealing is complete. It is further confirmed by XPS core-level spectra, shown in Figure S2 in the Supporting Information, that in all three films the C 1s and the N 1s peaks are of the same intensity. To investigate the orientation of the CH3NH3 we use C 1s angle-dependent XAS; these spectra are shown in Figure 2. Here the terms “normal” and “grazing” refer to the incidence angle of the X-ray beam relative to the sample. The C 1s XAS spectra are similar to the spectra of related materials available in the literature,20 and we may tentatively assign the resonant LUMO feature at 285 eV to C−N σ* bonding and the resonant features between 288 and 291 eV to C−H σ* bonding. We find that the C 1s XAS spectra samples #2 and #3 exhibited significant angle dependence, while the spectra from sample #1

incident X-ray beam is at an angle φ to the sample surface normal n. The CH3NH3 are oriented such that their C−N bond axis z is parallel to the surface of the sample. The X-ray beam therefore is at an azimuthal angle θ = π − φ relative to z. If a horizontally polarized X-ray beam is incident at an angle of θ relative to the z axis, the X-ray spectrum I(E) is given by eq 1, where E is the energy, Ti,f(E) is the transition matrix element, and ρS(E) is the density of states with symmetry S.22 I(E) ∝ |T1s,p(E)|2 (sin 2 θρz (E) + cos2 θρx , y (E))

(1)

When the X-ray beam is close to normal incidence, as shown in Figure 3a, then θ ≈ π, which enhances the pz factor in eq 1. Analogously, when the X-ray beam is at grazing incidence, as shown in Figure 3b, then θ ≈ 0, and the px,y factor in eq 1 is enhanced. Simply put, X-ray transitions are enhanced when the 2864

dx.doi.org/10.1021/jz501472d | J. Phys. Chem. Lett. 2014, 5, 2863−2867

The Journal of Physical Chemistry Letters

Letter

unoccupied orbital is largely parallel to the electric field of the X-ray (shown as the wavy green line in Figure 3). If the CH3NH3 are parallel to the surface, as shown in Figure 3, then at normal incidence the electric field will be parallel to some of the z axes of the CH3NH3 molecules, greatly enhancing their contribution to the LUMO resonance in the XAS spectrum. The C 1s XAS spectra for samples #2 and #3 therefore suggest that the C−N bond axis is parallel to the surface of the sample. Previous studies have suggested that the CH3NH3 are disordered and even rotating at room temperature,10,11 although other studies stress the importance of hydrogen bonding with iodine.14,15 Our picture of the C−N axis of CH3NH3 as being parallel with the surface still allows a large degree of disorder, but we need to justify that our interpretation of the C 1s XAS is correct. It is important to note that XAS is not a probe of the unoccupied ground-state orbitals. The electronic structure probed by XAS is dominated by the “final state rule”, which has a hole in a core level. This hole reduces the screening of the nuclear charge, which tends to pull the local unoccupied orbitals to lower energies.22 Explicitly calculating the effect of the core hole on the ground-state electronic structure is a challenging task,23 so herein we use crystal field (CF) theory to approximate the influence of the core hole. CF is a very wellknown approximation for determining the electronic structure of ionic compounds, and although it has largely been supplanted by ligand field theory or tight-binding models it is an appropriate approximation for modeling the core hole effect on insulating materials.24 This is because the core hole contracts and localizes the unoccupied states and because the XAS transition involves populating an otherwise unoccupied orbital with a single electron. These two effects minimize electronic interactions, making CF an appropriate approximation to understanding the rough shape of the XAS spectrum. The unoccupied partial DOS for the C site in CH3NH3 is shown in Figure 4a. Our calculated DOS is essentially the same

We can calculate the energy shift of an orbital with symmetry S ,m (i.e., the usual orbital and magnetic quantum numbers) within CF theory by calculating the matrix elements for the perturbation caused by a neighboring site24 using eq 2, where qA and rA⃗ are the charges and positions of neighboring site A, respectively, and r ⃗ is the position of the electron. UmCF1, m2 =

∑− A

qAe 4πε0

Sm1

1 Sm 2 | r ⃗ − rA⃗ |

(2)

Because both the C and N sites in CH3NH3 have the same local C3v symmetry (here we are ignoring the lead−iodine lattice because it is has a much smaller influence on the local orbitals than the bonding neighbors), the Um1,m2CF matrix is diagonal in m1,m2, and is degenerate in m1 = m2 = ±1. Because the core-hole contracts the orbitals, we can safely assume that r ⃗ < r ⃗ A for all A. In this case, the shifts in the C p unoccupied orbitals due to the CF perturbation are given in eqs 3 and 4, where θ is the azimuthal angle between the H sites and the C− N bond axis. U zCF = −

3q ⎞ e ⎛ qN 2e + H⎟ − ⟨r 2⟩ ⎜ 4πε0 ⎝ rN rH ⎠ 20πε0

⎞ ⎛q 3q ⎜ N3 + H3 (3 cos2 θ − 1)⎟ 2rH ⎠ ⎝ rN UxCF ,y = −

(3)

3q ⎞ e ⎛ qN e + H⎟ + ⟨r 2⟩ ⎜ 4πε0 ⎝ rN rH ⎠ 20πε0

⎞ ⎛q 3q ⎜ N3 + H3 (3 cos2 θ − 1)⎟ 2rH ⎠ ⎝ rN

(4)

Using the calculated geometry and charges and approximating ⟨r2⟩, the square of half the shortest bond length (in this case (bH/2)2), the obtained CF energy shifts are given in Table 1. Table 1. CF Parameters and Energy Shifts Used to Model the XASa parameter

value

parameter

value

qN (e) rN (Å) qH (e) rH (Å)

0.887 1.49 −0.211 1.11

θ UzCF (eV) Ux,yCF (eV) ΔU (eV)

70.93° −1.12 +0.02 −1.13

a qX is the charge of the neighboring site X and rX is the bond length to the neighboring site X (neighboring sites are H or N). θ is the azimuthal angle of the H sites relative to the C−N axis, USCF is the CF energy shift for orbitals of symmetry S, and ΔU is the difference between UzCF and Ux,yCF.

Figure 4. Calculated unoccupied p DOS for C in CH3NH3 and how the XAS spectra were modeled using CF theory. (a) Calculated unoccupied ground-state C pz and px,y DOS. (The Fermi level is set to 0 eV.) (b) Approximating the core hole perturbed DOS as local orbitals centered on the lowest bands. (c) Applying the shift in the pz orbital relative to the px,y orbital calculated from CF theory. (d) Measured (from sample #3) and modeled C 1s XAS.

For the CF calculation, we use the Mulliken charges from the electronic structure calculation. It is well known that Mulliken charges are unsatisfactory for coordination chemistry because they are very dependent on the choice of basis set.26 However, the partial DOS is also very dependent on the choice of basis set, so for our purposes using Mulliken charges is more consistent than using one of the alternative, basis-independent charges. With the CF energy shifts from Table 1, we can now model the XAS from the ground-state DOS. We first approximate the pz,px,y DOS as two localized orbitals centered on the lowest available energy band from the ground-state DOS, as shown in

as those previously reported.10,15,25 The z axis is projected along the C−N bond. Looking at the ground-state unoccupied DOS, we see that the pz and px,y (by px,y we mean px + py here) are basically degenerate. This is not what we observe in the XAS spectra, and this shows the need to understand the effect of the core hole on the ground-state DOS. 2865

dx.doi.org/10.1021/jz501472d | J. Phys. Chem. Lett. 2014, 5, 2863−2867

The Journal of Physical Chemistry Letters

Letter

Figure 4b. The core hole will cause all orbitals to shift to lower energies,22 but within our model this shift is unknown. We therefore ignore the absolute energy scale and only shift the pz orbital by the appropriate amount relative to the px,y orbital. (See Table 1.) The resulting relative energies of the orbitals are shown in Figure 4c. We finally can compute the model XAS by weighting each orbital according to the incidence angle given in eq 1, adding an arctangent to model the absorption edge step, and providing some Gaussian broadening to mimic the experimental resolution. The comparison of the model XAS spectra with the measured XAS spectra for sample #3 is shown in Figure 4d. Although there is quite a bit of fine structure in the orbital resonances of the measured XAS spectra that our simple CF model cannot account for, the CF orbitals are in good agreement with features found in the measured XAS, and, most importantly, the CF model accurately reproduces the observed angular-dependence in the XAS. In particular, although the unoccupied C p orbitals are largely degenerate in the ground state, this model reveals why the core hole perturbation shifts these orbitals apart as observed in the measured XAS. Incidentally, our CF model accurately reproduces the core-hole perturbation in the N 1s XAS spectra (see the Supporting Information); however, we have omitted the N 1s XAS spectra from this Letter because they do not contribute anything more to our discussion of the alignment of CH3NH3. To summarize, in the combination of CF theory and DFT analysis, the angle-resolved C 1s XAS spectra suggest that the C−N axis of the CH3NH3 in thin films prepared by one-step deposition (samples #2 and #3) is parallel to the surface of the films, while the CH3NH3 are randomly oriented in films prepared by two-step deposition (sample #1). Furthermore, the degree of orientation of CH3NH3 in the thin films decreases as the film thickness increases, although we stress that the similar XRD patterns for all samples suggests that only a common plane for C−N bonding, not a common C−N bond direction. This finding provides a connection between the organic cations and the thin-film preparation methods. With the CH3NH3 aligned parallel to the surface of the films, any structural changes related to expansion or contraction of the unit cell axis perpendicular to the surface will be relatively unhindered by the CH3NH3. This could prove important to the performance of layered devices, especially if thermal expansion is significant. Alignment of the CH3NH3 parallel to the film surface could frustrate lattice matching to the adjacent layers. The planar alignment of the CH3NH3 still allows a variety of orientations relative to the iodine sites, and it is possible that there is substantial hydrogen bonding between NH3 and I. However, because there are I sites on every vertex of the pseudocubic organic framework containing the CH3NH3 and we observe alignment of CH3NH3 relative to the surface of the sample, it is unlikely that the hydrogen bonding is the only mechanism for this long-range order. It is not possible at present to meaningfully compare the device performance of perovskites prepared by one- and twostep deposition because the morphology of these films is quite different. However, as synthesis techniques become more refined and reliable, the alignment of the CH3NH3 dipoles could be important in tuning the optical properties of organometal perovskites.11 Our present work is therefore valuable by not only demonstrating CH3NH3 alignment in a particular preparation of CH3NH3PbI3 but also by providing a simple model for interpreting the C 1s XAS spectra CH3NH3

because angle-resolved XAS is a relatively simple and increasingly common tool for in situ device characterization.

■

EXPERIMENTAL METHODS The CH3NH3PbI3 thin films used in this study were prepared from two different synthesis strategies: one- and two-step deposition. In brief, for the two-step deposition we first spincoat a PbI2 film and then form CH3NH3PbI3 by soaking the PbI2 film in a solution containing CH3NH3I. For the one-step deposition, a mixed solution of CH3NH3I and PbI2 was spincoated on to the substrate, followed by annealing at 100 °C for 5 min under dry N2. The films were characterized with XRD and scanning electron microscopy (SEM). We performed XAS measurements at beamline BL24A at the National Synchrotron Radiation Research Center (NSRRC) in Hsinchu, Taiwan.27 The samples were mounted on a stainlesssteel sample holder, and the thin-film side was facing the incident beam at various angles. The measurements were performed in surface-sensitive total-electron yield (TEY) mode using fast scan (100 s/spectrum). Electronic structure calculations were performed with the SIESTA density functional theory (DFT) code.28,29 The crystal structure and detailed parameters used in the calculation can be found in the Supporting Information. Following previous theoretical approaches, we started with a “pseudo-cubic” cell and performed a full geometry relaxation.30 Full details of the preparation, characterization, and calculation methods can be found in the Supporting Information.

■

ASSOCIATED CONTENT

* Supporting Information S

Detailed description of thin-film synthesis, SEM characterization, X-ray spectroscopy measurements, core-level XPS spectra of the thin films, angle-dependent XAS spectra at the C 1s, and details of the DFT calculation. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (J.A.M.). *E-mail: [email protected] (L.L.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by FUNSOM, Soochow-Western Center for Synchrotron Radiation Research. B.S. acknowledges the National Basic Research Program of China (973 Program) (2012CB932402). We thank Dr. Yaw-Wen Yang and Dr. ChiaHsin Wang for their technical support at the NSRRC.

■

REFERENCES

(1) Mitzi, D. B. Synthesis, Structure, and Properties of OrganicInorganic Perovskites and Related Materials. Prog. Inorg. Chem. 1999, 48, 1−121. (2) Mitzi, D. B.; Feild, C. A.; Harrison, W. T. A.; Guloy, A. M. Conducting Tin Halides with a Layered Organic-based Perovskite Structure. Nature 1999, 369, 467−469. (3) Papavassiliou, G. C. Synthetic Three- and Lower-dimensional Semiconductors based on Inorganic Units. Mol. Cryst. Liq. Cryst. 1996, 296, 231−238.

2866

dx.doi.org/10.1021/jz501472d | J. Phys. Chem. Lett. 2014, 5, 2863−2867

The Journal of Physical Chemistry Letters

Letter

Band of K3:Anthracene and K3:Phenanthrene. J. Phys. Chem. C 2013, 117, 19616−19621. (24) Cho, D.; Oh, S.; Kim, D.; Tanaka, A.; Park, J. Investigation of Local Symmetry Effects on the Electronic Structure of Manganites: Hexagonal YMnO3 versus Orthorhombic LaMnO3. Phys. Rev. B 2009, 79, 035116. (25) 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/CH3NH3PbI3Perovskite Solar Cell Interfaces. J. Phys. Chem. Lett. 2014, 5, 648−653. (26) Guerra, C. F.; Handgraaf, J. W.; Baerends, E. J.; Bickelhaupt, F. M. Voronoi Deformation Density (VDD) Charges: Assessment of the Mulliken, Bader, Hirshfeld, Weinhold, and VDD Methods, for Charge Analysis. J. Comput. Chem. 2003, 25, 189−210. (27) Lai, L. J.; Tseng, P. C.; Yang, Y. W.; Chung, S. C.; Song, Y. F.; Cheng, N. F.; Chen, C. C.; Chen, C. T.; Tsang, K. L. Current Status of the Wide-Range (10−1500 eV) Spherical Grating Monochromator Beamline at SRRC. Nucl. Instrum. Methods Phys. Res., Sect. A 2001, 467−468, 586−588. (28) Ordejon, P.; Artacho, E.; Soler, J. M. Self-Consistent Order-N Density-Functional Calculations for Very Large Systems. Phys. Rev. B 1996, 53, R10441−R10444. (29) Soler, J. M.; Artacho, E.; Gale, J. D.; Garc, A.; Junquera, J.; Ordejon, P.; Danial, S. P. SIESTA Method for Ab Initio Order-N Materials. J. Phys.: Condens. Matter 2002, 11, 2745−2779. (30) 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.

(4) Colella, S.; Mosconi, E.; Fedeli, P.; Listorti, A.; Gazza, F.; Orlandi, F.; Ferro, P.; Besagni, T.; Rizzo, A.; Calestani, G.; et al. MAPbI3‑xClx Mixed Halide Perovskite for Hybrid Solar Cells: The Role of Chloride as Dopant on the Transport and Structural Properties. Chem. Mater. 2013, 25, 4613−4618. (5) 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. (6) Kim, H. S.; Lee, C. R.; Im, J. H.; Lee, K. B.; Moehl, T.; Marchioro, A.; Moon, S. J.; Humphry-Baker, R.; Yum, J. H.; Moser, J. E.; et al. Lead Iodide Perovskite Sensitized All-solid-state Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (7) Liu, D.; Kelly, T. L. Perovskite Solar Cells with a Planar Heterojunction Structure prepared using Room-Temperature Solution Processing Techniques. Nat. Photonics 2013, 8, 133−138. (8) Liu, M.; Johnston, M. B.; Snaith, H. J. Efficient Planar Heterojunction Perovskite Solar Cells by Vapour Deposition. Nature 2013, 501, 395−398. (9) 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. (10) Baikie, T.; Fang, Y.; Kadro, J. M.; Schreyer, M.; Wei, F.; Mhaisalkar, S. G.; Graetzel, M.; White, T. J. Synthesis and Crystal Chemistry of the Hybrid Perovskite (CH3NH3)PbI3 for Solid-state Sensitised Solar Cell Applications. J. Mater. Chem. A 2013, 1, 5628. (11) Brivio, F.; Walker, A. B.; Walsh, A. Structural and Electronic Properties of Hybrid Perovskites for High-efficiency Thin-film Photovoltaics from First-principles. APL Mater. 2013, 1, 042111. (12) Poglitsch, A.; Weber, D. Dynamic Disorder in Methylammoniumtrihalogenoplumbates (II) observed by Millimeter-wave Spectroscopy. J. Chem. Phys. 1987, 87, 6373. (13) Onoda-Yamamuro, N.; Yamamuro, O.; Matsuo, T.; Suga, H. PT Phase Relations of CH3NH3PbX3 (X=Cl, Br, I) Crystals. J. Phys. Chem. Solids 1992, 53, 277−281. (14) Mitzi, D. B. Templating and Structural Engineering in Organic− Inorganic Perovskites. J. Chem. Soc., Dalton Trans. 2001, 1−12. (15) 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. (16) Burschka, J.; Pellet, N.; Moon, S. J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Gratzel, M. Sequential Deposition as a Route to High-performance Perovskite-sensitized Solar Cells. Nature 2013, 499, 316−319. (17) Bi, D.; Moon, S.-J.; Häggman, L.; Boschloo, G.; Yang, L.; Johansson, E. M. J.; Nazeeruddin, M. K.; Grätzel, M.; Hagfeldt, A. Using a Two-step Deposition Technique to Prepare Perovskite (CH3NH3PbI3) for Thin Film Solar Cells based on ZrO2 and TiO2 Mesostructures. RSC Adv. 2013, 3, 18762. (18) Kasi, G. K.; Dollahon, N. R.; Ahmadi, T. S. Fabrication and Characterization of Solid PbI2 Nanocrystals. J. Phys. D: Appl. Phys. 2007, 40, 1778−1783. (19) Zheng, Z.; Liu, A.; Wang, Y.; Li, Z.; Lau, W. M.; Zhang, L. In Situ Growth of Epitaxial Lead Iodide Films Composed of Hexagonal Single Crystals. J. Mater. Chem. 2005, 15, 4555−4559. (20) Lasky, P.; Lu, P.; Yang, M.; Osgood, R. M., Jr.; Bent, B.; Stevens, P. NEXAFS Measurements Indicating a Tilted Molecular Orientation for Methyl Halides on GaAs (110). Surf. Sci. 1995, 336, 140−148. (21) Lindau, I.; Spicer, W. The Probing Depth in Photoemission and Auger-Electron Spectroscopy. J. Electron Spectrosc. Relat. Phenom. 1974, 3, 409−413. (22) Hahner, G. Near Edge X-Ray Absorption Fine Structure Spectroscopy as a Tool to Probe Electronic and Structural Properties of Thin Organic Films and Liquids. Chem. Soc. Rev. 2006, 35, 1244− 1255. (23) Pitman, A.; McLeod, J.; Khozeimeh Sarbisheh, E.; Kurmaev, E.; Muller, J.; Moewes, A. X-Ray Spectroscopic Study of the Conduction 2867

dx.doi.org/10.1021/jz501472d | J. Phys. Chem. Lett. 2014, 5, 2863−2867