Highly Stable New Organic–Inorganic Hybrid 3D Perovskite

Sep 24, 2018 - NUS Graduate School for the Integrative Sciences and Engineering, National University of Singapore, Singapore 117456 , Singapore...
0 downloads 0 Views 4MB Size
Letter Cite This: J. Phys. Chem. Lett. 2018, 9, 5862−5872

pubs.acs.org/JPCL

Highly Stable New Organic−Inorganic Hybrid 3D Perovskite CH3NH3PdI3 and 2D Perovskite (CH3NH3)3Pd2I7: DFT Analysis, Synthesis, Structure, Transition Behavior, and Physical Properties Xixia Liu,† Nengduo Zhang,†,‡ Baoshan Tang,† Mengsha Li,† Yong-Wei Zhang,§ Zhi Gen Yu,*,§ and Hao Gong*,† †

Department of Materials Science and Engineering, National University of Singapore, Singapore 117576, Singapore NUS Graduate School for the Integrative Sciences and Engineering, National University of Singapore, Singapore 117456, Singapore § Institute of High Performance Computing, Singapore 138632, Singapore

J. Phys. Chem. Lett. 2018.9:5862-5872. Downloaded from pubs.acs.org by KAOHSIUNG MEDICAL UNIV on 10/05/18. For personal use only.



S Supporting Information *

ABSTRACT: The feasibility of Pd-based organic−inorganic hybrid perovskites is comprehensively explored with both theoretical and experimental methods for the first time. Experimentally, the new 3D perovskite CH3NH3PdI3 (tetragonal, I4cm) can be transited to a new 2D perovskite (CH3NH3)3Pd2I7 (tetragonal, P4mm) by modulating the ratio of the organic part to inorganic part. The structure, lattice parameters, and symmetry of these two perovskites are verified by a series of simulations, refinement, and characterizations. The basic optical and electronic properties of these two new perovskites are characterized and calculated with DFT for future applications. Interestingly, both types of perovskites exhibit long stability in air with 50% relative humidity. Two-day stability for the 3D perovskite and one-week stability for the 2D perovskite are observed, consistent with our DFT calculation that 2D perovskite (CH3NH3)3Pd2I7 is more energetically stable than 3D hybrid perovskite CH3NH3PdI3.

O

wells for both electrons and holes.7 In the multiquantum well layered perovskites, the sizable organic cations are more favorable to self-assemble than the organic materials with small chains. However, organic components with large size build high barriers in the multiquantum well structure, thus diminishing the optoelectronic performance. Under this consideration, it is attractive to explore the possibility of using small organic components, such as MA and FA, to synthesize the layered hybrid perovskite compounds. Currently, the well-known and widely reported OIHPs are mainly limited to group IV A elements-based composites, in particular, for lead (Pb) and tin (Sn).10,11 The Pb-based perovskite compounds exhibit superior optical and electronic properties, especially in the solar cell field with certified efficiency over 22%.3,12 However, the poor stability and high toxicity of Pb-based perovskites hinder their commercialization.13 With substitution of Pb, less toxic Sn-based perovskite compounds have attracted significant attention. Despite the lower toxicity of Sn versus Pb, Sn-based perovskite materials have worse stability than Pb-based ones because Sn2+ is easily oxidized into Sn4+ after exposure to air.14 Therefore, exploring another divalent metal-based perovskite with nontoxicity and high stability is desirable. It is well-known that hybrid perovskite compounds are primarily formed using divalent

rganic−inorganic hybrid perovskite materials (OIHPs) have been universally investigated during recent years because of their attractive properties and promising potential applications such as light-emitting diodes (LEDs),1 solar cells,2,3 lasers,4 detectors,5 and transistors.6 The superior properties and their far-ranging applications mainly originate from their unique compositions and structures. OIHPs are formed by hybridizing the organic components, such as methylammonium (MA), formamidinium (FA), and phenylethylammonium (PEA), with inorganic compounds like PbI2 and SnI2 into one composite. Generally, depending on the size of organic cations, 3D perovskites are based on smaller organic components (MA, FA, etc.) with a chemical formula of AMX3, and layered (quasi-2D) perovskites are based on large organic molecules (PEA, AEI, etc.) with chemical formula [(R− NH3)2An−1MnX3n+1],7 where A, M, and X are monovalent organic cations (R is an organic group), divalent metal cations, and halide anions, respectively. The structure of 3D perovskite is analogous to the conventional perovskite CaTiO3.8 The 2D perovskite structure can be regarded as a combination of alternative stacking of organic layers and inorganic octahedron layers along a specific direction.9 Due to the difference in the composition and structure, 2D perovskite materials usually exhibit disparate electronic, optical, and phase stability properties, in comparison with 3D perovskite counterparts. This is because 2D hybrid perovskites form the multiquantum well structure, where the inorganic layers with larger dielectric constants and smaller optical band gaps function as quantum © 2018 American Chemical Society

Received: August 16, 2018 Accepted: September 24, 2018 Published: September 24, 2018 5862

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

Figure 1. (a) θ−2θ XRD pattern of the synthesized film, grounded powder, and calculated pattern. (b) Pawley fit profile of the powder XRD pattern. The inset shows the structure of the 3D perovskite CH3NH3PdI3.

and octahedral factor μ are calculated as 0.924 and 0.391, respectively. The tolerance factor t lies in the range of 0.76− 1.13, and the octahedral factor μ is close to the range of 0.377−0.895,21 suggesting the formability of a new organic− inorganic hybrid palladium-based (Pd-based) perovskite compound CH3NH3PdI3. However, the 3D perovskite is perhaps not the most stable phase in terms of its lower octahedral factor μ value. Besides, as shown in the equations above, the concept of the tolerance factor and octahedral factor for a perovskite compound is a dimensionless number calculated merely from the ratios of ionic radii without considering the crystal chemistry of the divalent metals M2+, such as their preferred coordination and orbital interaction.22 Therefore, even though the concepts of tolerance factor and octahedral factor are designed for 3D perovskites, other phases such as layered perovskite (2D) can also be estimated when crystal chemistry is taken into consideration. Experimentally, a stoichiometric mixture of palladium iodide (PdI2) and methylammonium iodide (MAI) (atomic ratio of 1:1) was first dissolved in hydroiodic acid in the air atmosphere, following our previous feasible recipe for synthesizing a different perovskite compound (CH3NH3)2PdCl4.23 Unexpectedly, however, the XRD pattern (Figure S1a) suggested that the synthesized material after annealing was a mixture of precursors rather than a new compound as the X-ray diffraction (XRD) peaks matched with MAI and PdI2 (Figures S1a and S2), which may be due to the fact that the reaction atmosphere in HI is not suitable. Then we changed the reaction atmosphere by substituting HI with N,N-dimethylformamide (DMF) and conducting the synthesis under a N2 atmosphere while maintaining other reaction conditions unchanged. The XRD pattern of the film formed under this new condition exhibited a different set of peaks (Figure S1b), indicating that the precursors could now react with each other by using this new recipe and synthesis conditions. However, the reaction among precursors was incomplete. Therefore, we further modified the reaction conditions and synthesized films by spin-coating the mixed precursor solutions several times, followed by annealing the spin-coated films at 80 °C. Detailed experimental methods for the synthesis of the new perovskite compounds are described in the Supporting Information. Both energy-dispersive X-ray spectroscopy (EDX) and X-ray photoelectron spectroscopy (XPS) were employed to

cations M2+ at the central site, which constrains the possibility for synthesizing new perovskite compounds, but not all divalent metal cations can be assembled into the perovskite structure. For example, the Zn2+-, Cd2+-, Co2+-, and Hg2+based composites are tetrahedrally coordinated rather than octahedrally coordinated in perovskite compounds.15,16 To the best of our knowledge, all of the reported OIHPs using MA as the organic component are summarized in Table S1. As a rule of all reported perovskites, the band gaps of perovskite compounds decrease with the change of halide elements from Cl to I, exhibiting superior photoelectric properties.17,18 Therefore, hybrid perovskite compounds using I as the halide elements, such as MAPbI3 and MASnI3, are the dominating materials in photoelectric fields. Compared with the summarized reports shown in Table S1, the Pd-based hybrid perovskites previously reported by our group possess higher stability and lower toxicity. Apart from that, it is theoretically predicted and experimentally verified that incorporating Pd into the conventional OIHP MAPbI3 to form the mixed MAPdxPb1−xI3 perovskites can decrease the band gap by over 22% and enhances light harvesting. However, the specific structure and properties of MAPdI3 perovskite are not well explored yet.19 Considering these findings and existing issues, it is extremely urgent and appealing to explore the new 3D or 2D hybrid perovskite compounds based on PdI6 octahedra and investigate their stability. Motivated by the summarization and analysis of the reported perovskite, in this work, we experimentally and theoretically studied new Pd-based OIHPs and their chemical and physical information. Because no structural data on perovskite CH3NH3PdI3 can be found, our first task is to analyze whether a perovskite CH3NH3PdI3 can be synthesized and remain stable. Therefore, first, the formability and stability of the aimed possible perovskite is estimated using the Goldschmidt tolerance factor t and octahedral factor20,21 t = (rA + rX)/ 2 (rM + rX)

(1)

μ = rM /rX

(2)

where the rA, rM, and rC are the ionic radii for A, M, and X ions in the general perovskite formula AMX3, respectively. For a possible perovskite CH3NH3PdI3, the ionic radii values of CH3NH3+, Pd2+, and I−, corresponding to the A, M, and X site, are 0.180, 0.086, and 0.22 nm, respectively. By substituting these ionic radii values into eqs 1 and 2, the tolerance factor t 5863

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters determine the composition ratios of synthesized films. EDX (Figure S3a) reveals atomic percentages of 33.94% (C), 24.52% (N), 10.14% (Pd), and 31.40% (I) for this sample, corresponding to the normalized atomic ratio of C:N:Pd:I = 3.347:2.418:1.000:3.096. The carbon contamination and detection limitation of EDX characterization for light elements give rise to inaccurate ratios for C and N.24 The atomic ratio of Pd:I = 1.000:3.096 is similar to that of the possible 3D perovskite CH3NH3PdI3. For XPS characterization, we obtained an atomic fraction of 25.57% (C), 14.26% (N), 14.75% (Pd), and 45. 43% (I) (see Figure S3b), giving an atomic ratio of C:N:Pd:I = 1.734:0.967:1.000:3.080. As carbon contamination and surface sensitivity contribute to a higher amount of C for XPS characterization, the amount of carbon in the sample is generally overestimated when using XPS characterization. On the basis of the composition analysis, the atomic ratio of N:Pd:I approximately equals 1:1:3, which also corresponds to the composition of a possible 3D perovskite CH3NH3PdI3. Although the above elemental characterizations suggest the formation of CH3NH3PdI3, structural confirmation is also needed for the 3D perovskite. Figure 1a shows the XRD pattern of the obtained film and grounded powder. Several main peaks in the XRD pattern are located at 14.79, 21.11, 25.72, 29.79, 33.34, 42.56, 45.17, and 50.52°. The distribution of the XRD peaks is similar to the XRD pattern of the widely studied Pb-based hybrid perovskite compound MAPbI3.25,26 The similarity of the XRD pattern indicates an analogous tetragonal structure with a space group of I4cm (#140 in the International Space Group Table).27 On the basis of this understanding, the model of this new compound is built as shown in the inset of Figure 1b. The tetragonal structure of CH3NH3PdI3 is stacked by a cubic single cell with a size of 6.018 Å according to the symmetry of 4/mmm. Through modulation of lattice parameters to a = b = 8.511 Å and c = 12.020 Å, the calculated powder XRD pattern (as shown in the vertical line in Figure 1a and comparison of peak positions between the experimental and simulated XRD pattern in Table S2) is well-matched with its experimental XRD pattern. The fitted XRD pattern confirms that the obtained compound is a new 3D organic−inorganic perovskite with the chemical formula of CH3NH3PdI3, together with the composition analysis. In order to further verify the structure of this new perovskite compound, we conducted Pawley refinement. Pawley refinement is currently a universally used structural refinement method because it addresses the problem of leastsquares ill-conditioning of the Rietveld refinement due to overlapping reflections.28 Premised on the experimental powder XRD pattern and the predicted structure, the Pawley refinement result is shown in Figure 1b. The Pawley refined XRD pattern with optimized lattice parameters of a = b = 8.496 Å and c = 12.196 Å is well-fitted with the experimental XRD pattern, which reconfirms the calculated 4/mmm symmetry and lattice parameters of this new 3D hybrid perovskite compound. Our previous studies found that the Pd-based hybrid perovskite could self-assemble the layered structure for (CH 3 NH 3 ) 2 PdX 4 (X = Cl, Br) using small organic molecules.22,23 Therefore, another 1 mol of MAI was added in the 1 M 3D CH3NH3PdI3 perovskite precursor solution to explore the formability of a new laminar perovskite compound. After stirring and heating at 80 °C for 30 min, a brownish black film was formed, which is different from the black color of the

3D perovskite CH3NH3PdI3 film. The composition of the newly synthesized film was analyzed using EDX and XPS characterization. The atomic fractions of C, N, Pd, and I as revealed by EDX are 41.59, 27.57, 6.81, and 24.02% (as shown in Figure S4a), respectively, corresponding to an atomic ratio of 6.107:4.408:3.527:1.000 for C:N:Pd:I. The overestimation of C and N stems from the technique limitation. XPS characterization manifests atomic fractions of 30.18% (C), 16.92% (N), 11.43% (Pd), and 41.47% (I), which is normalized to an atomic ratio of 2.640:1.480:1.000:3.628 for C: N: Pd: I (as shown in Figure S4b). The normalized atomic ratio of this newly synthesized compound is shown in Table 1 Table 1. Elemental Ratios of Synthesized Perovskite Compoundsa elemental ratios characterizations MAPdI3

MA3Pd2I7

a

EDX XPS theoretical EDX XPS theoretical

C

N

I

Pd

3.347 1.734 1 6.107 2.640 1.5

2.418 0.967 1 4.048 1.480 1.5

3.096 3.080 3 3.527 3.628 3.5

1 1 1 1 1 1

The ratios are normalized to Pd.

together with the 3D perovskite CH3NH3PdI3. On the basis of the composition analysis, the atomic ratio of N:Pd:I is close to 1.5:1.0:3.5 or 3.0:2.0:7.0. This atomic ratio is different from our previous reports for layered perovskites (CH3NH3)2PdCl423 or (CH3NH3)2PdBr422 with an atomic ratio of 2.0:1.0:4.0 for N:Pd:X (X = Cl or Br). This suggests that the synthesized compound is not the normal layered perovskite composed of a single layer of inorganic component consecutively connected with another single layer of organic component. Mitzi et al. reported that the general chemical formula for the layered perovskite analogue is (R− NH3)2An−1MnX3n+1, where R−NH3 and A are organic components.7 Thus, the chemical formula of the new synthesized compound is likely to be the layered perovskite with n = 2, rather than n = 1, according to the compositional analysis. To determine the structure of this newly achieved “layered perovskite”, XRD characterization was conducted. The XRD pattern of the synthesized compound is shown in Figure 2a. Compared with the film XRD pattern of the 3D perovskite CH3NH3PdI3 as presented in Figure 1a, the XRD pattern of the newly synthesized compound is very different, with several obvious periodic peaks located at 9.91, 19.80, 25.44, 29.79, and 40.40°. In contrast to the precursor peaks in Figure S2, there are no emergent peaks that can be assigned to MAI peaks. This comparison suggests that all five peaks originate from a new compound and the added MAI plays a significant role in the transition from the 3D perovskite to this new compound. The periodic peaks distribution in Figure 2a is analogous to XRD patterns of other layered OIHPs.6,22,23,29,30 This distribution similarity in XRD patterns indicates the layered characteristic for the synthesized compound. To determine the specific crystal structure of this new layered perovskite analogue, we scratch the films from substrates and ground them into powder. As shown in the red curve in Figure 2b, more XRD peaks appear after grinding into powder, which facilitates the crystal structure determination. Compared with the powder 5864

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

Figure 2. (a) θ−2θ XRD pattern of the synthesized film. (b) Comparison of the powder XRD pattern between 3D perovskite CH3NH3PdI3 and 2D perovskite (CH3NH3)3Pd2I7. The blue vertical line shows the overlapped peaks between the 3D perovskite and 2D perovskite. The inset illustrates the feature of the layered perovskite with 3D perovskite characteristic. (c) Pawley fit profile of the powder XRD pattern. The inset shows the structure of the 2D perovskite (CH3NH3)3Pd2I7. Comparison of (d) the experimental SAED pattern and (e) simulated SAED pattern based on the proposed structure. The lattice parameters and symmetry of the gained 2D perovskite (CH3NH3)3Pd2I7 based on powder XRD calculation, Pawley refinement, and SAED calculation are listed in the bottom panels (d) and (e).

The good matching between the Pawley refined XRD curve and experimental XRD curve verifies the validity of the proposed structure of the layered (quasi-2D) perovskite (CH3NH3)3Pd2I7. Selected-area electron diffraction (SAED) was further carried out to confirm the symmetry and lattice parameter of the 2D OIHP. The film XRD pattern implies the layered flakes lying parallel to the substrate (as shown in Figure 2a). Therefore, the sample was prepared by drop-casting on a TEM grid directly instead of the glass substrate. During TEM imaging, the electron beam is incident to the [00l] direction of the sample. The SAED pattern is shown in Figure 2d. The spot pattern of the SAED pattern features characteristics of a single crystal in the selected area. Different positions of the sample were captured randomly and repeated several times, and the same pattern was obtained, which illustrates the homogeneous phase of the sample. The measured lattice parameters are a = b = 8.65 Å, the same as the XRD analysis results and Pawley refined value. The simulated electron diffraction pattern on the basis of the crystal structure determined above was also conducted using the commercial TEM simulation software MacTempas version 2.2.15 (as shown in Figure 2e). The simulated SAED pattern also shows both bright and weak spots. As reported in our previous paper,22 the bright diffraction spots in the reciprocal space are contributed by the heavy atoms M2+ and X− in perovskite materials (Pd2+ and

XRD pattern of CH3NH3PdI3 (see the black curve in Figure 2b), several peaks in the powder XRD pattern of the layered perovskite have identical positions with a few peaks of CH3NH3PdI3, which are labeled using the vertical lines shown in Figure 2b. The identical positions of some XRD peaks illustrate that the 3D and 2D perovskites have some common characteristics in crystal structures, which will be further discussed in the later part. On the basis of the XRD comparison and the chemical formula (CH3NH3)3Pd2I7 obtained from the composition analysis, we propose the structure as follows: Two adjacent inorganic components are separated by one single organic layer, which is analogous to all of the layered perovskite structure.31−33 The difference is that the inorganic component is composed of two connected inorganic layers, similar to truncation of the two layers in the 3D perovskite structure. The schematic illustration of this proposed structure is shown as the inset in Figure 2b. On the basis of this proposed structure (the insets in Figure 2b,c), the calculated powder XRD can be matched with the experimental XRD, as shown in Figure 2b and Table S3 by modulating the lattice parameters to a = b = 8.63 Å and c = 17.90 Å. The Pawley fit profile presented in Figure 2c indicates the tetragonal structure with the P4mm space group (#99 in the International Space Table27) and the refined lattice parameters of a = b = 8.65 Å and c = 17.78 Å. 5865

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters I− in this work), while the weaker diffraction spots originated from the light atoms of molecules CH3NH3+. The light atoms result in a smaller atomic scattering factor, thus generating the difference in intensities of the diffraction spots for (210) between the simulated pattern and experimental pattern.23,34 Despite the minor difference in the intensity of the weak spots, the symmetry of 4mm is clearly clarified by the comparison of experimental and simulated SAED patterns. The composition analysis and structure determination show that the organic part MAI plays an essential role in the formation and transition between these two types of perovskites. The reported work in Table S1 shows that either 3D perovskites or 2D perovskites can be synthesized using a simple and small organic molecule. Compared with the literature review, it is amazingly founded that the organic molecule MAI can form both the 3D perovskite and the 2D perovskite by assembling the PdI6 octahedron with MAI. A mechanism regarding the significant function of the MAI in the transition from the 3D perovskite CH3NH3PdI3 and the 2D perovskite (CH3NH3)3Pd2I7 is proposed and shown in Figure 3. When a stoichiometric ratio of the organic molecule and

inorganic component (MAI:PdI2 = 1:1) is mixed in suitable solvents, a 3D perovskite tends to be formed (eq 3). However, the formed perovskite is not in the most stable state because of its critical value of octahedral factor and higher chemical potential compared to our DFT calculation (described clearly in a later section). After adding excess MAI into the 3D perovskite system, in order to transform into the most energystable status, the 3D perovskite would be disassembled into a single-layer system. The disintegration process under the impact of excess MAI is akin to a scissor-cutting of the 3D perovskite system. Thus, this effect of excess MAI is named the scissor effect in this work. During annealing, the cut single layer would be reassembled and recrystallized into the layered perovskite (eq 4). This mechanism can rationally explain the similar lattice parameters of a and b and the same symmetry in the a−b plane of the 3D perovskite and 2D perovskite because of their structurally akin single layer. CH3NH3I + PdI 2 → CH3NH3PdI3

(3)

2CH3NH3PdI3 + CH3NH3I → (CH3NH3)3 Pd 2I 7

(4)

To further confirm the rationality of this proposed mechanism, Raman and Fourier infrared spectroscopy (FTIR) characterizations were performed, and the results are presented in Figure 4. Raman spectra of the 3D perovskite CH3NH3PdI3, the 2D perovskite (CH3NH3)3Pd2I7, and the inorganic precursor PdI2 are shown in Figure 4a, which reveals the internal molecular vibrational modes in the low-frequency region. The peak located at 148.27 cm−1 disappears in the Raman pattern of the perovskite, as compared with the precursor PdI2, which is mainly attributed to the formation of a perovskite compound.22 Besides, Raman peaks of the 3D perovskite are virtually identical to the counterpart of the 2D perovskite. This illustrates the similar Pd−I vibrational mode in the PdI6 octahedron of both 3D and 2D perovskites. FTIR characterization could provide information regarding internal vibrational modes of organic components in the OIHPs.35 The FTIR peaks of the perovskite as shown in Figure 4b are indexed according to the reported FTIR spectra of CH3NH3I (MAI).36 In the range of low frequency, both the 3D perovskite and the 2D perovskite compounds exhibit a similar N−H bend, C−N stretch, and C−H bend at 1625.78, 1464.53, and 1393.36 cm−1, respectively (highlighted by an orange

Figure 3. Schematic illustration of the proposed transition behavior from 3D perovskite CH3NH3PdI3 to 2D perovskite (CH3NH3)3Pd2I7.

Figure 4. (a) Raman spectra of the precursor PdI2, the 3D perovskite CH3NH3PdI3, and the 2D perovskite (CH3NH3)3Pd2I7. (b) FTIR spectra of the 3D perovskite CH3NH3PdI3 and the 2D perovskite (CH3NH3)3Pd2I7. The low-frequency and the high-frequency vibrations are highlighted using an orange circle and a blue rectangle, respectively. 5866

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

Figure 5. XPS spectra of the comparison of (a) C, (b) N, (c) Pd, and (d) I between the 3D perovskite CH3NH3PdI3 and the 2D perovskite (CH3NH3)3Pd2I7.

peaks of Pd and I are approximately located at the same positions for both types of perovskites, implying a similar chemical environment of these two elements in the 3D perovskite and the 2D perovskite because of the same PdI6 octahedron. Nevertheless, the peaks of C and N of the 2D perovskite (CH3NH3)3Pd2I7 are shifted to higher binding energy positions by 0.3 eV in contrast to that of the 3D perovskite CH3NH3PdI3. This shift suggests a change of the chemical environment of these two elements in the two perovskite compounds, which supports the scissor effect of MAI proposed above. The cross-sectional SEM images of the sample are shown in Figure S5, which displays the growth difference between the 3D perovskite and the 2D perovskite. The 3D perovskite shows a bulk feature, but the 2D perovskite exhibits plate morphology. The difference of the morphology stems from the different growth mechanism due to the scissor effect of the excess MAI. Figure 6a,b presents the optical absorption spectra for these two types of samples. The 3D perovskite CH3NH3PdI3 shows an absorption band edge at around 720 nm, corresponding to the optical band gap of ∼1.72 eV. The 2D perovskite (CH3NH3)3Pd2I7 exhibits two absorption peaks, where the absorption band edge at 700 nm corresponds to an optical band gap of ∼1.79 eV. The other absorption peak situated at 410 nm is caused by defects.42,43 The steady-state photoluminescence (PL) characterization (Figure 6c) indicates an emission peak at 716 nm for the 3D perovskite, close to the position of its band edge. In contrast, no obvious PL peak at the position of the band edge for the 2D perovskite is

circle). The similar vibrational mode in the low-frequency range mainly originates from the internal vibration of the organic component. Different from the low-frequency range, the FTIR peaks in the high-frequency range change greatly. In comparison with the 3D perovskite CH3NH3PdI3, there are two extra peaks located at 3105.88 and 2928.73 cm−1, corresponding to the N−H stretch and C−H stretch, respectively (highlighted by a blue rectangle). It is reported that the organic part is interconnected with the inorganic part via the hydrogen bond N−H−X (X = Cl, Br, and I) to form the perovskite structure.37 In a 3D perovskite, the organic component is tagged by the inorganic component tightly, whereas the organic component becomes relatively loose in the 2D perovskite, which allows for the vibrational modes of the C−H stretch and N−H stretch. The analogous Raman peaks indicate the same vibrational status in the PdI6 octahedron of both 2D and 3D perovskites, but the difference in FTIR spectra suggests a disparate vibrational mode in the MA. The Raman and FTIR spectra clearly describe the different perovskite style based on the same PdI6 octahedron, verifying the scissor effect of MAI in the transition from 3D perovskite CH3NH3PdI3 to 2D perovskite (CH3NH3)3Pd2I7. High-resolution XPS was performed to analyze the chemical state of the elements in the two types of OIHPs. The spectra of C, N, Pd, and I are shown in Figure 5, where the positions of all presented peaks are calibrated using the adventitious C 1s (284.8 eV). As compared with literature, the peaks of C 1s, N 1s, Pd 3d, and I 3d could be assigned to the corresponding elemental chemical states in perovskite compounds.38−41 The 5867

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

Figure 6. UV−vis spectra of (a) the 3D perovskite CH3NH3PdI3 and (b) the 2D perovskite (CH3NH3)3Pd2I7. The insets show the corresponding Tauc plots and the energy level diagrams. (c) Normalized steady-state PL spectra and (d) TGA curves of the 3D perovskite and the 2D perovskite.

Figure 7. Calculated electrical band structures based on the optimized atomic model of the 3D perovskite CH3NH3PdI3 with the I4cm space group (shown in the middle upper panel) and the 2D perovskite (CH3NH3)3Pd2I7 with the P4mm space group (shown in the middle lower panel). (a,b) GGA-PBE and HSE band structures of the 3D I4cm perovskite and (c,d) PBE and HSE band structures of the 2D P4mm perovskite, respectively. The light gray, purple, gray, light blue, and pink balls in the middle panel models represent Pd, I, C, N, and H atoms, respectively.

determined and plotted in the insets of Figure 6a,b. Besides, both types of perovskite compounds exhibit ultrahigh stability, as compared with several hours or minutes of stability of Pbbased or Sn-based perovskites (as shown in Figure S7a−d). The 3D perovskite is stable for 2 days, and the 2D perovskite

recorded, which suggests an indirect band gap characteristic. Ultraviolet photoelectron spectroscopy (UPS) was characterized to probe the electronic state of the two new perovskite compounds (Figure S6). Integrated with the band gap value gained from UV−vis spectra, the energy level diagrams are 5868

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

narrower. The calculated band gaps are shown in Figure S9. Here, for the 3D perovskite CH3NH3PdI3, the calculated band gap is decreased to 0.39 eV (rotated toward Pd) from 0.50 eV (unperturbed) using the GGA-PBE functional and is decreased to 0.74 eV (rotated toward Pd) from 0.99 eV (unperturbed) using the HSE functional; for the 2D perovskite (CH3NH3)3Pd2I7, the calculated band gap is slightly decreased to 0.51 eV (rotated toward Pd) from 0.57 eV (unperturbed) using the GGA-PBE functional and is decreased to 1.21 eV (rotated toward Pd) from 1.31 eV (unperturbed) using the HSE functional. On the basis of our calculated band gaps shown in Figure S9, we can see that the dipole effect on the band gap is very weak and the band gap difference is about 0.1 eV. Overall, the calculated band gap of the 2D perovskite (CH3NH3)3Pd2I7 is larger than that of the 3D perovskite CH3NH3PdI3. The band gap change due to the dipole effect of MA cations also can be visualized based on the localization of the valence band maximum (VBM) and the conduction band minimum (CBM). The calculated charge densities of the CBM and the VBM states of 3D perovskite CH3NH3PdI3 with and without the dipole effect are shown in Figure S10. In contrast, MA cations rotated toward Pd make the CBM and the VBM delocalize and further reduce the band gap. The same trend as that shown in Figure S11 can also be seen in the structure of the 2D perovskite (CH3NH3)3Pd2I7. Comparing the calculated charge densities of the CBM and the VBM states of the 3D perovskite CH3NH3PdI3 shown in Figure S10 and of the 2D perovskite (CH3NH3)3Pd2I7 shown in Figure S11, it can be seen that the 3D perovskite CH3 NH3 PdI3 has more delocalized charge density of the CBM and the VBM states than the 2D perovskite (CH3NH3)3Pd2I7. It is well-known that a more delocalized charge density results in a narrower band gap; therefore, the 2D perovskite (CH3NH3)3Pd2I7 has a wider band gap than the 3D perovskite CH3NH3PdI3. Experimentally, a similar phenomenon has been reported that decreasing the dimensionality of the perovskite from 3D structures can cause an increase in the band gap.33 In our experimental study, we found that the 2D perovskite of (CH3NH3)3Pd2I7 has better phase stability than the 3D perovskite of CH3NH3PbI3, and the phase transition from 3D to 2D occurred under the optimized experimental conditions. To gain an in-depth understanding on the phase stability and transition, we calculated the phase diagrams of the two proposed perovskites via a convex hull method.47 On the basis of the definition, the convex hull method represents the Gibbs free energy of the compounds at zero temperature. In the DFT calculation, the formation energy is estimated to be the Gibbs free energy (T = 0). The convex hull method consists of phases that have an energy lower than that of any other phase or a linear combination of phases at the respective compositions. The stability or convex hull distance is defined as

can remain stable for 1 week in air at room temperature with relative humidity of 50%. The moisture stabilities of the 3D perovskite (∼2 days) and 2D perovskite (∼1 week) are impressively high as compared with the several minutes of stability of MASnI344,45 and a few hours of stability of MAPbI3.46,47 In addition to air stability, thermal stability was also explored by thermogravimetric analysis (TGA) measurement. The TGA curves in Figure 6d reveal that both the 3D perovskite CH3NH3PdI3 and 2D perovskite (CH3NH3)3Pd2I7 are thermally stable up to 250 °C, comparable to other hybrid perovskites.48,49 All of these physical properties are attractive for possible applications in the future. To gain insights into the experimental results, we investigated electrical properties of the two new perovskite materials by DFT calculations. The optimized models are shown in the middle panel of Figure 7 by fixing their lattice constants (experimental values). For 3D perovskite CH3NH3PdI3, the optimized bond length between Pd and I (dPd−I) is ∼3.04 Å along the c direction and slightly reduces to ∼3.01 Å in the x−y plane. In contrast, the bond length of dPd−I is slightly shorter than the reported bond length dPb−I in CH3NH3PbI3, which is 3.16 Å.46 From the view of the bond length, we may consider that Pd has stronger chemical bonding with I than Pb does; thus, CH3NH3PdI3 may has better stability than CH 3 NH 3 PbI 3 . For the 2D perovskite (CH3NH3)3Pd2I7, the optimized value of dPd−I varies between 2.67 and 2.78 Å along the c direction due to the 2D layered structure and between 2.67 and 3.78 Å in the x−y plane. On the basis of the optimized models, as shown in the middle panel of Figure 7, we investigate their electrical properties, and the calculated band structures of the two materials are shown in Figures 7. Here, it can be seen that the calculated GGA-PBE direct band gap is 0.50 eV for the structure of I4cm, as shown in Figure 7a. However, when the structure converts to 2D P4mm from 3D I4cm, we find that the band structure characteristics are changed to those of an indirect band gap, and the band gap slightly increases to 0.57 eV (Γ−Υ), as shown in Figure 7c. The band gap value of the 2D perovskite is larger than that of the 3D perovskite, in accordance with the experimental results. It should be noted that GGA-PBE generally provides an underestimated band gap, in contrast to the values from experimental measurement, but can well describe the band structure characteristics, e.g., indirect or direct. We further conducted in-depth hybrid functional (HSE06) studies to investigate the band structures, and the results are shown in Figure 7b,d, corresponding to structures of I4cm and P4mm, respectively. Here, it can be seen that the HSE band gaps of 3D I4cm and 2D P4mm are increased to 0.99 and 1.31 eV, respectively. Not surprisingly, our calculated band gaps are smaller than our measured ones, and our DFT results reveal that the 3D perovskite of CH3NH3PbI3 has a direct band structure, while the 2D perovskite of (CH3NH3)3Pd2I7 has an indirect band structure, which are consistent with our experimental observations. The arrangement of cations might result in a dipole effect on hybrid perovskites. The spin polarization and the spin orbital coupling (SOC) were considered in our DFT calculations to explore the dipole effect on the electrical properties. To distinguish the MA dipoles’ effect on electrical properties, we rotated MA toward Pd (with the dipole effect), in contrast with the optimized model (unperturbed MA without the dipole effect). The calculated band structures are shown in Figure S8, in which we found that MA dipoles would make the band gap

C Hstab = HfC − Hf

(5)

where HCf is the formation energy of the compound C with different structures and Hf is the convex hull energy for the compound C. For a given phase C, the convex hull distance can be positive (unstable) or native (stable). The formation energy of each structure is determined with respect to the chemical potentials of pure elements. In our case, the formation energy of the hybrid perovskite is defined as 5869

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

CH3NH3PdI3 is formed with the tetragonal structure I4cm space group and lattice constants of a = b = 8.50 Å and c = 12.20 Å. The 2D perovskite (CH3NH3)3Pd2I7 is transformed with the tetragonal structure P4mm space group and lattice constants of a = b = 8.65 Å and c = 17.78 Å, after adding excess MAI into the 3D perovskite solution. The structure is confirmed by composition analysis, structure-based calculation, Pawley refinement, symmetry analysis, and DFT optimization. The transformation behavior is ascribed to the scissor effect of the added MAI for a more thermodynamically stable phase, as proved by the Raman, FTIR, and XPS characterizations and theoretical formation enthalpy calculation. The band gap and electronic structures of these two types of new perovskite compounds are characterized using UV−vis UPS and DFT calculations. The dipole effect of the MA cation is studied and shows weak influence on the optical properties in the two types of perovskite based on the results of DFT calculations. Interestingly, both perovskite compounds show superior stability (2 days for the 3D perovskite and 1 week for the 2D perovskite) as compared with the widely reported Pb- and Sn-based perovskites. The optimized structures with the calculated phase diagrams provide deep insight into the good stability of the two Pd-based hybrid perovskites. The higher stability and transition behavior may open a door to further understanding of the growth mechanism of perovskite materials and exploration of their suitable application in the future.

HfCH3NH3PdI3 = E(CH3NH3PdI3) − E(CH3NH3) − μPd − 3μI

(6)

Hf(CH3NH3)3Pd 2I7 = E((CH3NH3)3 Pd 2I 7) − 3E(CH3NH3) − 2μPd − 7μI

(7)

where E(CH3NH3PdI3) and E((CH3NH3)3Pd2I7) are the total energies of the two investigated perovskites. E(CH3NH3) is the total energy of CH3NH3 molecules, and μPd,I are the chemical potentials of Pd and I, respectively. For elemental chemical potentials, they can be obtained from DFT calculation of the total energies of their respective bulk structures. The calculated formation energies are −0.51 and −0.47 eV based on the above equations. According to our calculated results, we found that both types of perovskites have negative formation energies, which suggests that they are thermodynamically stable for synthesis. On the basis of the calculated formation energies, the phase stability of the two proposed perovskites via the convex hull method is shown in Figure 8. Here, the solid



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b02524. Experimental method, computation method, reported organic−inorganic perovskite, XRD patterns of precursor materials of CH3NH3I and PdI2, EDX and XRD characteristics, band structures and band gap, charge density of the CBM and the VBM states, and references (PDF)

Figure 8. Phase stability of the two proposed perovskites. The solid diamonds and solid triangle represent the formation energy of the 2D perovskite (CH3NH3)3Pd2I7 (Pd:I = 2:7) and the 3D perovskite CH3NH3PdI3 (Pd:I = 1:3), respectively. The convex hull passes through the blue and red dashed lines represent the stabilities of the two phases.



diamond and solid triangle represent the formation energy of the 2D perovskite (CH3NH3)3Pd2I7 (Pd:I = 2:7) and the 3D perovskite CH3NH3PdI3 (Pd:I = 1:3), respectively, and the hollow triangle represents the minimum formation energy to stabilize the structure for the 3D perovskite CH3NH3PdI3. The 3D I4cm perovskite lies above the convex hull formed by the 2D P4mm perovskite, which thermodynamically supports the transformation of the 3D perovskite into the 2D perovskite, consistent with our observation in the experiment. On the basis of eq 5, the convex hull distance is equal to the energy difference between the solid and hollow triangles, and the computed value is ∼0.12 eV. Our calculated results shown in Figure 8 suggest that the 2D perovskite (CH3NH3)3Pd2I7 is more stable than the 3D perovskite CH3NH3PdI3, and the former structure also has better long-term stability than the latter structure, which is consistent with our experimental results. Meanwhile, our theoretical results shown in Figure 8 can also provide insight into our experimental observation of their different stabilities (2 vs 7 day). In conclusion, two types of new perovskite materials based on the PdI6 octahedron and MAI organic component are reported for the first time. Experimentally, the 3D perovskite

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected] (H.G.). *E-mail: [email protected] (Z.G.Y.). ORCID

Zhi Gen Yu: 0000-0002-3718-6027 Hao Gong: 0000-0002-0410-5147 Author Contributions

X.L. and N.Z. carried out the sample growth, characteristics, and properties measurement. Z.G.Y. carried out the DFT calculations. All authors performed data analysis and manuscript writing. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the Singapore Ministry of Education Academic Research Fund Tier 2 MOE2016-T2-1049, Grant Number R284-000-157-112 and the China Scholarship Council 201606050114. Computational resources were provided by the A*STAR Computational Resource 5870

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters

Fernández-Lorenzo, C.; Martín-Calleja, J. The Impact of Pd on the Light Harvesting in Hybrid Organic-inorganic Perovskite for Solar Cells. Nano Energy 2017, 34, 141−154. (20) Kieslich, G.; Sun, S.; Cheetham, A. K. Solid-state Principles Applied to Organic−Inorganic Perovskites: New Tricks for an Old Dog. Chem. Sci. 2014, 5, 4712−4715. (21) Li, C.; Lu, X.; Ding, W.; Feng, L.; Gao, Y.; Guo, Z. Formability of ABX3 (X = F,Cl,Br,I) Halide erovskites. Acta Crystallogr., Sect. B: Struct. Sci. 2008, 64, 702−707. (22) Liu, X.; Huang, T. J.; Zhang, L.; Tang, B.; Zhang, N.; Shi, D.; Gong, H. Highly Stable New Organic-Inorganic Perovskite (CH3NH3)2PdBr4: Synthesis, Structure and Physical Properties. Chem. - Eur. J. 2018, 24, 4991−4998. (23) Huang, T. J.; Thiang, Z. X.; Yin, X.; Tang, C.; Qi, G.; Gong, H. (CH3NH3)2PdCl4: A Compound with Two-Dimensional OrganicInorganic Layered Perovskite Structure. Chem. - Eur. J. 2016, 22, 2146−2152. (24) Bell, D.; Garratt-Reed, A. Energy Dispersive X-ray Analysis in the Electron Microscope; Garland Science, 2003; Vol. 49. (25) 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− 5641. (26) Wang, Q.; Lyu, M.; Zhang, M.; Yun, J. H.; Chen, H.; Wang, L. Transition from the Tetragonal to Cubic Phase of Organohalide Perovskite: The Role of Chlorine in Crystal Formation of CH3NH3PbI3 on TiO2 Substrates. J. Phys. Chem. Lett. 2015, 6, 4379−4384. (27) Hahn, T. International Tables for Crystallography: Brief Teaching ed. of Vol. A, Space-group Symmetry; Kluwer Academic Publishing, 1996; Vol. 1. (28) Pawley, G. S. Unit-Cell Refinement From Powder Diffraction Scans. J. Appl. Crystallogr. 1981, 14, 357−361. (29) Mitzi, D. B. Synthesis, Structure, and Properties of Organicinorganic Perovskites and Related Materials. Progress in Inorganic Chemistry; John Wiley & Sons, Inc.: New York, 2007; Vol. 48, pp 1− 121. (30) Mitzi, D. B.; Prikas, M. T.; Chondroudis, K. Thin Film Deposition of Organic-inorganic Hybrid Materials Using a Single Source Thermal Ablation Technique. Chem. Mater. 1999, 11, 542− 544. (31) Mitzi, D. B.; Feild, C.; Harrison, W.; Guloy, A. Conducting Tin Halides with a Layered Organic-based Perovskite Structure. Nature 1994, 369, 467−469. (32) Kataoka, S.; Banerjee, S.; Kawai, A.; Kamimura, Y.; Choi, J. C.; Kodaira, T.; Sato, K.; Endo, A. Layered Hybrid Perovskites with Micropores Created by Alkylammonium Functional Silsesquioxane Interlayers. J. Am. Chem. Soc. 2015, 137, 4158−4163. (33) Smith, I. C.; Hoke, E. T.; Solis-Ibarra, D.; McGehee, M. D.; Karunadasa, H. I. A Layered Hybrid Perovskite Solar-cell Absorber with Enhanced Moisture Stability. Angew. Chem., Int. Ed. 2014, 53, 11232−11235. (34) Williams, D. B.; Carter, C. B. High Energy-Loss Spectra and Images. Transmission Electron Microscopy; Springer, 2009; pp 715− 739. (35) Glaser, T.; Muller, C.; Sendner, M.; Krekeler, C.; Semonin, O. E.; Hull, T. D.; Yaffe, O.; Owen, J. S.; Kowalsky, W.; Pucci, A.; et al. Infrared Spectroscopic Study of Vibrational Modes in Methylammonium Lead Halide Perovskites. J. Phys. Chem. Lett. 2015, 6, 2913− 2918. (36) Bhatt, P.; Pandey, K.; Yadav, P.; Tripathi, B.; P., C. K.; Pandey, M. K.; Kumar, M. Investigating the Charge Carrier Transport within the Hole-transport Material Free Perovskite Solar Cell Processed in Abient Air. Sol. Energy Mater. Sol. Cells 2015, 140, 320−327. (37) Mitzi, D. B. Templating and Structural Engineering in Organic−inorganic Perovskites. J. Chem. Soc., Dalton Trans. 2001, 0, 1−12.

Centre, Singapore (A*CRC) and the National Supercomputing Centre Singapore (NSCC).



REFERENCES

(1) Xiao, Z.; Kerner, R. A.; Zhao, L.; Tran, N. L.; Lee, K. M.; Koh, T.-W.; Scholes, G. D.; Rand, B. P. Efficient Perovskite Light-Emitting Diodes Featuring Nanometre-Sized Crystallites. Nat. Photonics 2017, 11, 108−115. (2) 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. (3) 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.; et al. Iodide Management in Formamidinium-Lead-Halide−Based Perovskite Layers for Efficient Solar Cells. Science 2017, 356, 1376−1379. (4) Tang, B.; Dong, H.; Sun, L.; Zheng, W.; Wang, Q.; Sun, F.; Jiang, X.; Pan, A.; Zhang, L. Single-Mode Lasers Based on Cesium Lead Halide Perovskite Submicron Spheres. ACS Nano 2017, 11, 10681−10688. (5) Saidaminov, M. I.; Adinolfi, V.; Comin, R.; Abdelhady, A. L.; Peng, W.; Dursun, I.; Yuan, M.; Hoogland, S.; Sargent, E. H.; Bakr, O. M. Planar-Integrated Single-Crystalline Perovskite Photodetectors. Nat. Commun. 2015, 6, 8724. (6) Kagan, C. R.; Mitzi, D. B.; Dimitrakopoulos, C. D. OrganicInorganic Hybrid Materials as Semiconducting Channels in Thin-Film Field-Effect Transistors. Science 1999, 286, 945−947. (7) Mitzi, D. B.; Chondroudis, K.; Kagan, C. R. Organic Inorganic Electronics. IBM J. Res. Dev. 2001, 45, 29−45. (8) De Graef, M.; McHenry, M. E. Structure of Materials: An Introduction to Crystallography, Diffraction and Symmetry; Cambridge University Press, 2012. (9) Cheng, Z. Y.; Lin, J. Layered Organic−Inorganic Hybrid Perovskites: Structure, Optical Properties, Film Preparation, Patterning and Templating Engineering. CrystEngComm 2010, 12, 2646− 2662. (10) Zhao, Y.; Zhu, K. Organic-Inorganic Hybrid Lead Halide Perovskites for Optoelectronic and Electronic Applications. Chem. Soc. Rev. 2016, 45, 655−689. (11) Polman, A.; Knight, M.; Garnett, E. C.; Ehrler, B.; Sinke, W. C. Photovoltaic Materials: Present Efficiencies and Future Challenges. Science 2016, 352, aad4424. (12) 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. (13) Liu, X.; Li, B.; Zhang, N.; Yu, Z.; Sun, K.; Tang, B.; Shi, D.; Yao, H.; Ouyang, J.; Gong, H. Multifunctional RbCl Dopants for Efficient Inverted Planar Perovskite Solar Cell with Ultra-High Fill Factor, Negligible Hysteresis and Improved Stability. Nano Energy 2018, 53, 567−578. (14) Hu, H.; Dong, B.; Zhang, W. Low-Toxic Metal Halide Perovskites: Opportunities and Future Challenges. J. Mater. Chem. A 2017, 5, 11436−11449. (15) Horiuchi, K.; Ishihara, H.; Terao, H. Differential Scanning Calorimetry Studies on Structural Phase Transitions and Molecular Motions in (CH3NH3)2MBr4 and [(CH3)2NH2]2MBr4 (M = Zn, Cd, and Hg). J. Phys.: Condens. Matter 2000, 12 (22), 4799−4806. (16) Roccanova, R.; Ming, W.; Whiteside, V. R.; McGuire, M. A.; Sellers, I. R.; Du, M. H.; Saparov, B. Synthesis, Crystal and Electronic Structures, and Optical Properties of (CH3NH3)2CdX4 (X = Cl, Br, I). Inorg. Chem. 2017, 56, 13878−13888. (17) Huang, L. Y.; Lambrecht, W. R. L. Electronic Band Structure, Phonons, and Exciton Binding Energies of Halide Perovskites CsSnCl3, CsSnBr3, and CsSnI3. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 88, 165203. (18) Dimesso, L.; Hamburger, M.; Jaegermann, W.; Dimamay, M. Properties of CH3NH3PbX3 (X = I, Br, Cl) Powders as Precursors for Organic/Inorganic Solar Cells. Chem. Mater. 2014, 26, 6762−6770. (19) Navas, J.; Sánchez-Coronilla, A.; Gallardo, J. J.; Piñero, J. C.; De los Santos, D.; Martín, E. I.; Hernández, N. C.; Alcántara, R.; 5871

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872

Letter

The Journal of Physical Chemistry Letters (38) Xie, H.; Liu, X.; Lyu, L.; Niu, D.; Wang, Q.; Huang, J.; Gao, Y. Effects of Precursor Ratios and Annealing on Electronic Structure and Surface Composition of CH3NH3PbI3 Perovskite Films. J. Phys. Chem. C 2016, 120, 215−220. (39) Philippe, B.; Park, B.-W.; Lindblad, R.; Oscarsson, J.; Ahmadi, S.; Johansson, E. M. J.; Rensmo, H. Chemical and Electronic Structure Characterization of Lead Halide Perovskites and Stability Behavior under Different ExposuresA Photoelectron Spectroscopy Investigation. Chem. Mater. 2015, 27, 1720−1731. (40) Grasso, V.; Silipigni, L. X-ray Photoemission Spectra and X-ray Excited Auger Spectrum Investigation of the Electronic Structure of Pd3(PS4)2. J. Vac. Sci. Technol., A 2003, 21, 860−865. (41) Kumar, G.; Blackburn, J. R.; Albridge, R. G.; Moddeman, W. E.; Jones, M. M. Photoelectron Spectroscopy of Coordination Compounds. II. Palladium Complexes. Inorg. Chem. 1972, 11, 296−300. (42) ElBatal, F.; Selim, M.; Marzouk, S.; Azooz, M. UV-vis Absorption of the Transition Metal-doped SiO2−B2O3−Na2O Glasses. Phys. B 2007, 398, 126−134. (43) Srikant, V.; Clarke, D. R. On the Optical Band Gap of Zinc Oxide. J. Appl. Phys. 1998, 83, 5447−5451. (44) Nishikubo, R.; Ishida, N.; Katsuki, Y.; Wakamiya, A.; Saeki, A. Minute-Scale Degradation and Shift of Valence-Band Maxima of (CH3NH3)SnI3 and HC(NH2)2SnI3 Perovskites upon Air Exposure. J. Phys. Chem. C 2017, 121, 19650−19656. (45) Liao, Y.; Liu, H.; Zhou, W.; Yang, D.; Shang, Y.; Shi, Z.; Li, B.; Jiang, X.; Zhang, L.; Quan, L. N.; et al. Highly Oriented LowDimensional Tin Halide Perovskites with Enhanced Stability and Photovoltaic Performance. J. Am. Chem. Soc. 2017, 139, 6693−6699. (46) Aristidou, N.; Sanchez-Molina, I.; Chotchuangchutchaval, T.; Brown, M.; Martinez, L.; Rath, T.; Haque, S. A. The Role of Oxygen in the Degradation of Methylammonium Lead Trihalide Perovskite Photoactive Layers. Angew. Chem. 2015, 127, 8326−8330. (47) Smecca, E.; Numata, Y.; Deretzis, I.; Pellegrino, G.; Boninelli, S.; Miyasaka, T.; La Magna, A.; Alberti, A. Stability of SolutionProcessed MAPbI3 and FAPbI3 Layers. Phys. Chem. Chem. Phys. 2016, 18, 13413−13422. (48) Dang, Y.; Liu, Y.; Sun, Y.; Yuan, D.; Liu, X.; Lu, W.; Liu, G.; Xia, H.; Tao, X. Bulk Crystal Growth of Hybrid Perovskite Material CH3NH3PbI3. CrystEngComm 2015, 17, 665−670. (49) Zheng, Y.-Y.; Wu, G.; Deng, M.; Chen, H.-Z.; Wang, M.; Tang, B.-Z. Preparation and Characterization of A Layered Perovskite-Type Organic−Inorganic Hybrid Compound (C8NH6− CH2CH2NH3)2CuCl4. Thin Solid Films 2006, 514, 127−131.

5872

DOI: 10.1021/acs.jpclett.8b02524 J. Phys. Chem. Lett. 2018, 9, 5862−5872