Roles of SnX2 (X = F, Cl, Br) Additives in Tin-Based Halide

Sep 27, 2018 - Preserving the stability of Sn-based halide perovskites is a primary concern in developing photovoltaic light-absorbing materials for l...
0 downloads 0 Views 4MB Size
Download PDF
Letter Cite This: J. Phys. Chem. Lett. 2018, 9, 6024−6031

pubs.acs.org/JPCL

Roles of SnX2 (X = F, Cl, Br) Additives in Tin-Based Halide Perovskites toward Highly Efficient and Stable Lead-Free Perovskite Solar Cells Jin Hyuck Heo,†,⊥ Jongseob Kim,‡,⊥ Hyungjun Kim,§ Sang Hwa Moon,† Sang Hyuk Im,*,† and Ki-Ha Hong*,∥ †

Department of Chemical and Biological Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, Korea Samsung Electronics Co., Ltd, 130 Samsung-ro, Yeongtong-gu, Suwon-si, Gyeonggi-do 16678, Korea § Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), Yuseong-gu, Daejeon 34141, Korea ∥ Department of Materials Science and Engineering, Hanbat National University, 125 Dongseo-daero, Yuseong-Gu, Daejeon, 34158, Korea

Downloaded via UNIV OF TEXAS AT EL PASO on October 29, 2018 at 19:07:20 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

‡

S Supporting Information *

ABSTRACT: Preserving the stability of Sn-based halide perovskites is a primary concern in developing photovoltaic light-absorbing materials for lead-free perovskite solar cells. Whereas the addition of SnX2 (X = F, Cl, Br) has been demonstrated to improve the photovoltaic performance of Sn halide perovskite solar cells, the mechanistic roles of SnX2 in the performance enhancement have not yet been studied appropriately. Here we perform a comparative study of CsSnI3 films and devices and examine how SnX2 additives affect their stability, and the results are corroborated by first-principles-based theoretical calculations. Unlike the conventional belief that the additives annihilate defects, we find that the additives effectively passivate the surface and stabilize the perovskite phase, promoting the stability of CsSnI3. Our mechanism suggests that SnBr2, which shows ca. 100 h of prolonged stability along with a high power conversion efficiency of 4.3%, is the best additive for enhancing the stability of CsSnI3.

H

The relatively slow improvement in the PCE of Sn-based PV cells is entirely related to the stability of the Sn halide PVKs. Whereas efficient CsSnI3 solar cells can be made with the Bγ phase of CsSnI3 (black phase), the Bγ phase easily converts to the Y phase (yellow color, band gap of 2.6 eV) when exposed to air or organic solvents.11 Moreover, Sn2+ can be easily oxidized to Sn4+, which is considered one of the main demerits of Sn-based halide PVKs, whereas Pb-based halide PVKs have only the Pb2+ state. Because of this reaction, the Bγ phase usually degrades to Cs2SnI6 (0D phase) in ambient air, and the product shows a ten times smaller absorption coefficient than the Bγ phase.12 In addition to the phase stability, the hole concentration of Bγ-CsSnI3 is too high for use as a light-harvesting material.13 The high hole concentration of CsSnI3 was ascribed to Sn vacancy (VSn) formation through density functional theory (DFT) calculations.13 To enhance the PCE of Sn PVK-based solar cells, the issues with both the phase stability and the high free carrier density should be resolved. Recently, the addition of SnX2 (X = F, Cl, Br, I) was reported to enhance the stability and efficiency of CsSnI3 solar cells.5,12,14,15 Kumar et al. reported a PCE of 2.02% for CsSnI3

alide perovskites (PVKs) are the most attractive lightabsorbing materials for emerging solar cells. The power conversion efficiency (PCE) of halide PVK solar cells has already exceeded 22% in 2016.1 The first application of PVKs in solar cells was performed by the Miyasaka group in 2009 as sensitizers for dye-sensitized solar cells,2 and stabilized solidstate PVK solar cells were reported in 2012 by the Park group3 and the Snaith group.4 However, well-known issues remain regarding the commercialization of PVK solar cells beyond laboratory-scale production. The use of ecofriendly light-absorbing materials is one of the mandatory requirements of photovoltaic cells, whereas the best PCE of PVK cells has been achieved by only Pb-based halide PVKs. As one of best alternatives to Pb, Sn-based PVKs of the form ASnI3 [A = Cs, CH3NH3+ (methylammonium, MA), CH(NH2)2+ (formamidinium, FA)], where Sn is in the same group as Pb in the periodic table, have a more suitable band gap for solar cells (1.25 to 1.40 eV)5 according to the Shockley−Queisser limit.6 The first renowned work on Sn-based PVK photovoltaic (PV) cells was reported by Kanatzidis and coworkers in 2014, in which the cells based on MASnBrxI1−x demonstrated a PCE of ∼5.7%.7 Noel et al. also achieved a PCE of >6% with MASnI3 in 2014.8 FA-based Sn perovskite solar cells have shown improved PCE. Ke et al. presented 7.14% of PCE with {en}FASnI3,9 and Shao et al. recently reported a PCE of 9% through the formation of 2D/3D hybrid FASnI310 © 2018 American Chemical Society

Received: August 19, 2018 Accepted: September 27, 2018 Published: September 27, 2018 6024

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters

Figure 1. Lattice structures of (a) Bγ-CsSnI3 (unit cell), (b) Y-CsSnI3 (1 × 2 × 1 supercell), and (c) cubic Cs2SnI6 (unit cell). Green, gray, and purple spheres represent Cs, Sn, and I, respectively. Light-purple polyhedra represent the octahedra made from the bonding of Sn and I.

samples protected by a N2 atmosphere and not protected by a N2 atmosphere, as shown in Figure 2. To separate the phase stability from oxidation by air or water vapor, we stored the CsSnI3 PVK films containing the SnX2 additive in a glovebox charged with N2 gas for 100 h. The corresponding X-ray diffraction (XRD) patterns are shown in Figure 2a−c. All samples exhibited a pure Bγ-CsSnI3 phase in the initial stage irrespective of the kind of additive. However, the CsSnI3 PVK films containing SnF2 and SnCl2 additives exhibited additional Y-CsSnI3 phases after 100 h of storage, as shown in Figure 2a,b. This result indicates that the SnF2 and SnCl2 additives cannot fully prevent the transformation of the pure Bγ-CsSnI3 film to Y-CsSnI3. In contrast, the CsSnI3 PVK film containing the SnBr2 additive maintained its initial Bγ-CsSnI3 phase for 100 h of storage in the glovebox, as shown in Figure 2c. To examine the air stabilities of the CsSnI3 PVK films containing the SnX2 additives, we collected the XRD patterns of the samples immediately after removal from the glovebox and after storage in air for 1 h, as shown in Figure 2d−f. All samples exhibited Y-CsSnI3 and Cs2SnI6 phases. Note that the Cs2SnI6 phase is formed by the oxidation of CsSnI3 because Sn4+ is more stable than Sn2+. These model experiments clearly indicate that we can prevent the formation of Cs2SnI6 by protecting the film with a N2 atmosphere, and thus inhibition of the transition to the Y phase is a fundamental requirement for enhancing the stability of Bγ-CsSnI3. Considering the phase stability of the N2-protected films, SnBr2 shows an outstanding ability to prevent the formation of the Y phase. The promising behaviors of SnBr2 addition can be also found in the surface scanning electron microscopy (SEM) images shown in Figure S1. The surface SEM images of the CsSnI3 perovskite films with SnF2 and SnCl2 additive exhibited phase segregation of additives because the aggregated particles were presented in the film. On the contrary, the CsSnI3 perovskite thin film with SnBr2 additive did not show severe aggregated particles in the film. The lattice constant changes by the addition of SnX2 are checked by comparing (202) peak positions which are summarized in Table S1. The addition of SnF2, SnCl2, and SnBr2 shifted (202) peak to lower diffraction angles by 0.04°, 0.05°, and 0.06° respectively from that of the single crystalline phase. The 0.06° shift corresponds to 0.2% lattice expansion, assuming the same lattice symmetry. However, it should be noted that the lattice constants of thin-film perovskites like our samples are quite sensitive to fabrication procedures and

solar cells via the addition of SnF2, in which the researchers argued that the addition of SnF2 caused virtually no replacement of I in the lattice with F because the lattice constants changed very little and SnF2 might be uniformly mixed in CsSnI3 based on X-ray photoelectron spectroscopy measurements.14 Marshall et al. demonstrated that the performance of CsSnI3 solar cells could be improved by codepositing PVK precursors with SnX2 (X = F, Cl, Br, I)12 and argued that SnCl2 preferentially segregates at the surface based on the angle dependency of the Cl 2p peaks in the high-resolution X-ray photoelectron spectrum. In addition to CsSnI3, SnX2 (X = F, Cl, Br) addition was adopted to make stable and efficient FASnI3-based solar cells.16,17 The roles of the added SnX2 in the phase stabilization and performance enhancement of Sn-based halide PVKs are still vague despite consecutive successes in making lead-free PVK solar light-absorbing materials. Kumar et al. argued that VSn formation was suppressed by the enhanced formation energy achieved by increasing the chemical potential of Sn, which was attributed to SnF2 addition through calculations.14 However, the measured hole density was still too high (>1017/cm3), and the calculated formation energy did not quantitatively match the experimental results. According to Marshall et al.’s results,12 SnBr2 and SnCl2 may also be effective in making efficient solar cells. This implies that SnX2 plays a critical role in phase stabilization or defect control in Sn-based halide PVKs. In this report, the role of SnX2 additives in the stability and efficiency of Sn-based halide PVK solar cells is studied through a combination of DFT calculations and experiments. On the basis of both experimental and theoretical investigation of the thermodynamic features, we conclude that SnBr2 is the best choice for preventing the phase transformation of black CsSnI3 PVK to the yellow phase. We considered three kinds of lattice structures, orthorhombic Bγ-CsSnI3, Y-CsSnI3 (yellow phase), and cubic Cs2SnI6, which are represented in Figure 1, based on prior reports on CsSnI3. Bγ-CsSnI3 shows 3D SnI6 octahedra aligned through corner sharing and has the best band gap of 1.3 eV for photovoltaic materials, whereas Y-CsSnI3 shows a 1D double-chain structure and has a band gap that is too wide for use as an active layer in solar cells.18 Bγ-CsSnI3 is transformed to Y-CsSnI3 when exposed to air for 1 h, and prolonged exposure to air or oxidation conditions converts Y-CsSnI3 to Cs2SnI6.18 We compared the roles of SnX2 (X = F, Cl, and Br) additives in the stability of films by measuring the phase changes of 6025

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters

Figure 2. (a−c) Phase and (d−f) air stabilities of CsSnI3 PVK films containing (a,d) SnF2, (b,e) SnCl2, and (c,f) SnBr2 as additives. Samples a−c were protected by and stored in a N2 atmosphere for 100 h, and samples d−f were removed from the N2 atmosphere and stored in the ambient atmosphere for 1 h.

model CsSnI3 PVK solar cells to determine whether the SnBr2 additive could enhance the device stability, as shown in Figure 3. We fabricated planar-type model devices composed of F-doped tin oxide (FTO) glass/∼50 nm thick dense blocking TiO2/ ∼500 nm thick CsSnI3 with 10 mol % SnX2 additive/∼30 nm thick poly triarylamine (PTAA)/∼60 nm thick Au and kept them in a glovebox charged with N2 gas. Figure 3a shows the external quantum efficiency (EQE) spectra of the PVK solar cells with SnX2 additive, which indicate that there was no significant difference between the EQE values of all samples. Figure 3b shows the current density−voltage (J−V) curves of the CsSnI3 PVK solar cells with the SnX2 additive. The model PVK solar cell with SnF2 additive exhibited an open-circuit voltage (Voc) of 0.41 V, a short-circuit current density (Jsc) of 18.0 mA/cm2, a fill factor (FF) of 46.3%, and a PCE of 3.40%. The devices with SnCl2 and SnBr2 as additives exhibited PCEs of 3.90 and 4.30%, respectively. The photovoltaic parameters of each sample are summarized in Table 1. Finally, we examined the stabilities of the N2-protected model CsSnI3 PVK solar cells

thermal stresses so that careful examination is necessary to determine precise lattice constants. The enhancement in the phase stability achieved by adding SnBr2 can also be observed in the corresponding UV−visible absorption spectra (Figure S2). We compared the UV−visible absorption spectra of N2-protected samples every 20 h for 100 h, as shown in Figure S2a−c, to confirm the phase stabilities of the CsSnI3 PVK films containing SnX2 additives. Similar to the XRD results, the absorption spectra of the CsSnI3 PVK films with SnF2 and SnCl2 as additives changed during storage due to the formation of Y-CsSnI3, whereas the absorption spectrum of the PVK film with SnBr2 additive did not change during storage. The absorption spectra of the unprotected samples exhibited rapid degradation, as shown in Figure S2d−f, due to rapid oxidation. However, the CsSnI3 PVK film with SnBr2 additive showed slower degradation than the films with SnF2 and SnCl2 as additives. The improvement in stability achieved by adding SnBr2 can also be observed in the device performance. We fabricated 6026

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters

Figure 3. (a) External quantum efficiency (EQE) spectra, (b) current density−voltage (J−V) curves, and (c) stabilities of the encapsulated planartype CsSnI3 PVK solar cells with SnF2, SnCl2, and SnBr2 additives. (d−f) corresponding J−V curves of the CsSnI3 PVK solar cells with SnX2 additives at the initial (0 h) and final (100 h) stage: (d) SnF2, (e) SnCl2, and (f) SnBr2.

energy of Y-CsSnI3 is slightly lower than that of Bγ-CsSnI3 by 6 meV/atom, and Y-CsSnI3 is preferred over Bγ-CsSnI3 under a volume of 44 Å3/atom. The density of Bγ-CsSnI3 is lower than those of Y-CsSnI3 and Cs2SnI6 (see Table S2), and the volume per atom of CsSnI3 is smaller than that of Cs2SnI6 (Figure S3a). Furthermore, the estimation of the stable phase region of CsSnI3 illustrates that the presence of abundant Sn can be one of the important factors in preventing the formation of Cs2SnI6 (Figure S3b). We studied two plausible roles of SnX2 in improving the device stability through DFT calculations. The first role is defect control. Kumar et al. argued that a reduction in the free carrier concentration is critical for making efficient photovoltaic light-absorbing materials and that vacancy formation can be suppressed by increasing the chemical potential of Sn (μSn).14 The other role is surface passivation; that is, SnX2 can protect the absorber layer by forming protective layers.12 Although the researchers presented a qualitative estimation through a bulk energy calculation in which the vacancy formation energy increased at higher μSn, more precise computation is necessary to reveal the role of the SnF2 additive in vacancy formation. To determine the stability of SnF2-related defect configurations in Bγ-CsSnI3, the formation energies of various configurations were calculated and are summarized in Table 2. (The calculation details can be found in the SI, and the atomic configurations can be found in Figure S4.) On the basis of the calculated formation energies, SnF2 prefers to occupy interstitial sites in the Bγ-CsSnI3 matrix. Whereas F may interact with I through interstitial binding, the formation energy of a F interstitial (Fi) increases with higher concentrations of additive, and SnF2 incorporation into VSn (2Fi (Sn share)) is not more preferred than the formation of a SnF2 interstitial (SnF2(i)).

Table 1. Summary of the Photovoltaic Properties of PlanarType CsSnI3 PVK Solar Cells with SnX2 Additives device SnF2

SnCl2

SnBr2

best initial (0 h) final (10 h) best initial (0 h) final (10 h) best initial (0 h) final (10 h)

Voc (V)

Jsc (mA/cm2)

FF (%)

PCE (%)

0.41 0.40 0.30 0.43 0.43 0.39 0.44 0.44 0.44

18.0 18.0 13.6 17.4 17.3 14.5 18.5 18.5 18.3

46.3 47.0 30.1 52.3 51.7 31.9 52.9 52.5 52.3

3.40 3.38 1.23 3.90 3.85 1.80 4.30 4.27 4.21

with SnX2 additive over time in ambient air (25 °C, 40% relative humidity), as shown in Figure 3c. The corresponding J−V curves of the CsSnI3 PVK solar cells with SnX2 additives in the initial stage (0 h) and final stage (10 h) are shown in Figure 3d−f. As expected from the results of XRD and UV−visible absorption spectroscopy, the CsSnI3 PVK solar cell with SnBr2 additive exhibited excellent performance stability, whereas the performance of the other cells quickly degraded over time. The role of SnX2 in the stability enhancement was investigated with DFT calculations using the Vienna ab initio simulation package (VASP),19,20 and the computational details are described in the SI. DFT calculations have been widely used to search for new Pb-free hybrid PVK candidates,21−25 and our group has also applied DFT to develop new compositions and study the electronic properties of hybrid PVKs.26,27 The experimental and calculated lattice constants are summarized in Table S2. Our DFT calculations demonstrate that Y-CsSnI3 is energetically preferred over Bγ-CsSnI3. The formation 6027

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters Table 2. Formation Energies of SnF2-Related Defects, a F Interstitial (Fi), Dual F Interstitials near One Another ((2Fi (near)) and far from One Another (2Fi (long)), SnF2 Incorporated into a Sn Vacancy (VSn) (2Fi (Sn share)), and a SnF2 Interstitial (SnF2(i))

formation energy (eV)

Fi

2Fi (long)

2Fi (near)

2Fi (Sn share)

SnF2(i)

0.14

0.50

0.45

1.06

−0.50

The effects of the SnF2(i) additive on VSn formation were evaluated by comparing the formation energy changes. Four different configurations of VSn were considered by changing the distance between VSn and SnF2(i) in the Bγ-CsSnI3 matrix. Table 3 shows that the SnF2(i) additive in Bγ-CsSnI3 Table 3. Sn Vacancy (VSn) Formation Energies When There Are No Other Inclusions (No Inclusion), a F Interstitial (Fi), Dual F Interstitials near One Another (2Fi (near)) and far from One Another (2Fi (long)), and a SnF2 Interstitial (SnF2(i))

formation energy (eV)

no inclusion

Fi

2Fi (long)

2Fi (near)

SnF2(i)

1.17

1.42

1.46

1.51

1.05−1.22

does not as efficiently prevent VSn formation as Fi, which is contrary to the common expectation. The density of states (DOS) of bulk Bγ-CsSnI3 and F- or SnF2-incorporated Bγ-CsSnI3 are presented in Figure 4. As with various Pb-halide PVKs,28 VSn does not form deep defect states and induces p-type doping characteristics, as shown in Figure 4b. The DOS of SnF2(i), as shown in Figure 4c, is located far from the band edge states so that the conduction band minimum (CBM) and the valence band maximum (VBM) of Bγ-CsSnI3 are not affected by SnF2(i). The DOS of VSn together with SnF2(i) in the Bγ-CsSnI3 supercell shows a similar feature to the DOS of VSn only (Figure 4d). Whereas Fi can reduce VSn formation, Fi can act as source of holes such as VSn, as shown in Figure 4e, and minimal Fi can form in Bγ-CsSnI3, considering its positive formation energy. As a result, our calculation shows that the defect annihilation induced by SnX2 additives may not be the principle origin of the stability enhancement of CsSnI3 solar cells. The DOS changes by SnBr2 and Br interstitials are calculated and represented in Figure S5, which shows that SnBr2 can play a similar role as SnF2. The projected DOS of SnBr2(i) does not contribute to front orbitals, and the DOS of VSn together with SnBr2(i) in the Bγ-CsSnI3 supercell show a similar feature as the DOS of VSn also. Bri acts as the source of holes like VSn (see Figure S5c). Surface or grain boundary (GB) passivation is a plausible origin of the stability improvement caused by the addition of SnX2 into CsSnI3. Marshall et al. argued that CsSnI3 could be stabilized by the formation of passivation layers made of SnX2.12 To study the effect of surface passivation by the Sn halide additives on the electronic structures, Sn halide adsorption on the Bγ-CsSnI3 slab was considered, as shown in Figure 5a. Sn halide adsorption on the (001̅) surfaces of SnF2, SnCl2, and SnBr2 is shown in Figure 5a. Because of the relative differences in electronegativity and ionic radius of X with respect to those of I, the adsorption energy decreases in the order of F, Cl, and Br. Negative adsorption energies indicate that passivation of the surface of Bγ-CsSnI3 by SnX2 is

Figure 4. Density of states (DOS) of (a) bulk Bγ-CsSnI3 and Bγ-CsSnI3 with (b) VSn, (c) SnF2(i), (d) VSn-SnF2(i), and (e) Fi. The DOS of CsSnI3 and SnF2(i)/Fi are represented by red and green, respectively. The calculated Fermi energy is set to zero.

energetically favorable. One plausible explanation for how SnX2 additives enhance the stability of Bγ-CsSnI3 can be found in the DOS figures shown in Figure 5b−d. As with the prior theoretical study on the surface of MAPbI3,29 the surface states do not make gap states within the band gap. Type-I band alignment is observed between SnX2 and Bγ-CsSnI3, in which the unoccupied states of SnX2 are located above the CBM of Bγ-CsSnI3 and the occupied states of SnX2 lie below the VBM of Bγ-CsSnI3. Thus the stronger binding of SnBr2 on the surface relative to that of the other SnX2 additives can improve the stability of Bγ-CsSnI3 by forming passivation layers. The changes in the energy difference between Y-CsSnI3 and Bγ-CsSnI3 induced by the addition of SnX2 can be another source of phase stabilization. The DFT-calculated energies suggest that the most stable phase is Y-CsSnI3, which 6028

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters

Figure 6. Total energy difference between Y-CsSnI3 and Bγ-CsSnI3 per formula unit (left y axis) and the solution energies (right y axis) with regard to SnX2 inclusion in the supercell (one SnX2 unit in a supercell containing 160 atoms). A negative energy difference means that Bγ-CsSnI3 is preferred over Y-CsSnI3.

solution energies of SnX2 in Bγ-CsSnI3 were determined to be positive for X = Cl and Br due to the relatively larger molecular volumes of those atoms. It is thus inferred that the incorporation of solubilizing additives such as a pyrazine can enhance the stability of Sn-halide PVKs by enhancing the solubility of SnCl2 or SnBr2.17 In conclusion, we studied the roles of SnX2 (X = F, Cl, Br) additives on the stabilization of Sn halide PVKs through DFT calculations and comparative experiments. The stability of CsSnI3 solar cells can be enhanced by adding SnX2 (X = F, Cl, Br), and SnBr2 shows the most promising behavior in terms of an extended stability and a high efficiency (4.3% PCE). Although SnX2 addition is commonly accepted to help prevent VSn formation in CsSnI3, our theoretical work shows that the relationship between VSn formation and SnX2 addition is negligible. Instead, the origin of the performance enhancement induced by SnX2 can be ascribed to surface passivation and stabilization of Bγ-CsSnI3. Our DFT results further demonstrate that SnBr2 is the most effective among the additives in passivating the surface (with the largest adsorption energy) and stabilizing the Bγ phase, which effectively corroborates our experimental results. We anticipate that our results will offer new insight into the mechanistic roles of stabilizing additives in Sn-based halide PVKs, which is useful for designing highly efficient and stable Pb-free halide PVK solar cells.

Figure 5. (a) Adsorption energies of SnF2, SnCl2, and SnBr2 on the (001̅) surface of Bγ-CsSnI3. The relaxed atomic structures are represented in the inset. Total density of states (DOS) (red lines) and projected DOS of Sn (blue) and halides (green) of adsorbed SnX2 (X = (b) F, (c) Cl, and (d) Br).

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.8b02555. Details of computation and experiments. Figure S1. SEM surface images of CsSnI3 perovskite thin films SnF2, SnCl2, and SnBr2 additives. Figure S2. Figure S2. UV− visible absorption spectra of encapsulated and unencapsulated CsSnI3 perovskite films containing SnF2, SnCl2, and SnBr2 additive with respect to the storage time. Figure S3. Formation energy changes with volume change of Bγ-CsSnI3, Y-CsSnI3, and Cs2SnI6. Phase diagram of Bγ-CsSnI3 and Cs2SnI6 with regard to μSn and μI. Figure S4. Atomic structures of Bγ-CsSnI3 supercells containing a F interstitial, dual F interstitials of which distance is near and distant, SnF2 incorporated into VSn, and SnF2 interstitial. Figure S5. Density of states of SnBr2(i), VSn-SnBr2(i), and Bri. Table S1. (202)

effectively explains the experimentally observed phase change of Bγ-CsSnI3 to Y-CsSnI3. The addition of SnX2 into the interstitial sites can enhance the phase stability of Bγ-CsSnI3 over other polymorphs and change the most stable thermodynamic phase from Y-CsSnI3 to Bγ-CsSnI3. Total energy calculations were carried out on (2 × 2 × 2) and (2 × 4 × 1) supercells of Y-CsSnI3 and Bγ-CsSnI3, respectively (both supercells contained 160 atoms). Then, the total energy differences between Y-CsSnI3 and Bγ-CsSnI3 were calculated and normalized to the number of formula units, as shown in Figure 6. Without an additive, the formation energy of Y-CsSnI3 is lower than that of Bγ-CsSnI3 (as shown in Figure 6 and Figure S2a). With SnF2, Y- and Bγ-CsSnI3 show comparable energetic stabilities. With SnCl2 and SnBr2, Bγ-CsSnI3 is effectively stabilized over Y-CsSnI3, albeit the additive concentration is low, as less than one SnX2 unit is included in the simulation cell consisting of 160 atoms. However, the 6029

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters

■

Based Hybrid Perovskite Solar Cells with 9% Efficiency. Adv. Energy Mater. 2018, 8 (4), 1702019. (11) da Silva, E. L.; Skelton, J. M.; Parker, S. C.; Walsh, A. Phase Stability and Transformations in the Halide Perovskite CsSnI3. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91 (14), 144107−144112. (12) Marshall, K. P.; Walker, M.; Walton, R. I.; Hatton, R. A. Enhanced Stability and Efficiency in Hole-Transport-Layer-Free CsSnI3 Perovskite Photovoltaics. Nat. Energy 2016, 1, 16178. (13) Chung, I.; Song, J.-H.; Im, J.; Androulakis, J.; Malliakas, C. D.; Li, H.; Freeman, A. J.; Kenney, J. T.; Kanatzidis, M. G. CsSnI3: Semiconductor or Metal? High Electrical Conductivity and Strong Near-Infrared Photoluminescence From a Single. J. Am. Chem. Soc. 2012, 134 (20), 8579−8587. (14) Kumar, M. H.; Dharani, S.; Leong, W. L.; Boix, P. P.; Prabhakar, R. R.; Baikie, T.; Shi, C.; Ding, H.; Ramesh, R.; Asta, M.; et al. Lead-Free Halide Perovskite Solar Cells with High Photocurrents Realized Through Vacancy Modulation. Adv. Mater. 2014, 26 (41), 7122−7127. (15) Sabba, D.; Mulmudi, H. K.; Prabhakar, R. R.; Krishnamoorthy, T.; Baikie, T.; Boix, P. P.; Mhaisalkar, S.; Mathews, N. Impact of Anionic Br − Substitution on Open Circuit Voltage in Lead Free Perovskite (CsSnI3‑XBrX) Solar Cells. J. Phys. Chem. C 2015, 119 (4), 1763−1767. (16) Liao, W.; Zhao, D.; Yu, Y.; Grice, C. R.; Wang, C.; Cimaroli, A. J.; Schulz, P.; Meng, W.; Zhu, K.; Xiong, R.-G.; et al. Lead-Free Inverted Planar Formamidinium Tin Triiodide Perovskite Solar Cells Achieving Power Conversion Efficiencies Up to 6.22%. Adv. Mater. 2016, 28 (42), 9333−9340. (17) Lee, S. J.; Shin, S. S.; Kim, Y. C.; Kim, D.; Ahn, T. K.; Noh, J. H.; Seo, J.; Seok, S. I. Fabrication of Efficient Formamidinium Tin Iodide Perovskite Solar Cells Through SnF2−Pyrazine Complex. J. Am. Chem. Soc. 2016, 138 (12), 3974−3977. (18) Kontos, A. G.; Kaltzoglou, A.; Siranidi, E.; Palles, D.; Angeli, G. K.; Arfanis, M. K.; Psycharis, V.; Raptis, Y. S.; Kamitsos, E. I.; Trikalitis, P. N.; et al. Structural Stability, Vibrational Properties, and Photoluminescence in CsSnI3 Perovskite Upon the Addition of SnF2. Inorg. Chem. 2017, 56 (1), 84−91. (19) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 47 (1), 558. (20) Kresse, G.; Furthmüller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6 (1), 15−50. (21) Debbichi, L.; Lee, S.; Cho, H.; Rappe, A. M.; Hong, K.-H.; Jang, M. S.; Kim, H. Mixed Valence Perovskite Cs2Au2I6: A Potential Material for Thin-Film Pb-Free Photovoltaic Cells with Ultrahigh Efficiency. Adv. Mater. 2018, 30, 1707001−1707006. (22) Yang, D.; Lv, J.; Zhao, X.; Xu, Q.; Fu, Y.; Zhan, Y.; Zunger, A.; Zhang, L. Functionality-Directed Screening of Pb-Free Hybrid Organic−Inorganic Perovskites with Desired Intrinsic Photovoltaic Functionalities. Chem. Mater. 2017, 29 (2), 524−538. (23) Chakraborty, S.; Xie, W.; Mathews, N.; Sherburne, M.; Ahuja, R.; Asta, M.; Mhaisalkar, S. G. Rational Design: a High-Throughput Computational Screening and Experimental Validation Methodology for Lead-Free and Emergent Hybrid Perovskites. ACS Energy Lett. 2017, 2 (4), 837−845. (24) Volonakis, G.; Filip, M. R.; Haghighirad, A.-A.; Sakai, N.; Wenger, B.; Snaith, H. J.; Giustino, F. Lead-Free Halide Double Perovskites via Heterovalent Substitution of Noble Metals. J. Phys. Chem. Lett. 2016, 7 (7), 1254−1259. (25) Jung, Y.-K.; Lee, J.-H.; Walsh, A.; Soon, A. Influence of Rb/Cs Cation-Exchange on Inorganic Sn Halide Perovskites: From Chemical Structure to Physical Properties. Chem. Mater. 2017, 29 (7), 3181− 3188. (26) Hong, K.-H.; Kim, H.; Debbichi, L.; Kim, H.; Im, S. H. Band Gap Engineering of Cs3Bi2I9 Perovskites with Trivalent Atoms Using a Dual Metal Cation. J. Phys. Chem. C 2017, 121 (1), 969−974. (27) Jeon, J.; Eom, T.; Lee, E.; Kim, S.; Kim, S.; Hong, K.-H.; Kim, H. Polymorphic Phase Control Mechanism of Organic−Inorganic

peak shift in the XRD results. Table S2. Lattice constants of Bγ-CsSnI3, Y-CsSnI3, and Cs2SnI6. (PDF)

AUTHOR INFORMATION

Corresponding Authors

*K.-H.H.: E-mail: [email protected]. *S.H.I.: E-mail: [email protected]. ORCID

Hyungjun Kim: 0000-0001-8261-9381 Sang Hyuk Im: 0000-0001-7081-5959 Ki-Ha Hong: 0000-0001-7457-8196 Author Contributions ⊥

J.H.H. and J.K. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was supported by National R&D Program through the National Research Foundation of Korea (NRF) (NRF-2015M1A2A2055836, NRF-2018R1A2B6007888, NRF-2017M3A7B4041698) and New & Renewable Energy Core Technology Program of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) (No. 20183010013820). Supercomputing resources including technical support were supported by the Supercomputing Center/ Korea Institute of Science and Technology Information (KSC2017-C2-0038).

■

REFERENCES

(1) 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 (6345), 1376−1379. (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 (17), 6050−6051. (3) Kim, H.-S.; Lee, C.-R.; Im, J.-H.; Lee, K.-B.; Moehl, T.; Marchioro, A.; Moon, S.-J.; Humphry-Baker, R.; Yum, J.-H.; Moser, J. E.; et al. Lead Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (4) Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J. Efficient Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide Perovskites. Science 2012, 338 (6107), 643−647. (5) Konstantakou, M.; Stergiopoulos, T. A Critical Review on Tin Halide Perovskite Solar Cells. J. Mater. Chem. A 2017, 5 (23), 11518− 11549. (6) Shockley, W.; Queisser, H. J. Detailed Balance Limit of Efficiency of P-N Junction Solar Cells. J. Appl. Phys. 1961, 32 (3), 510−511. (7) Hao, F.; Stoumpos, C. C.; Cao, D. H.; Chang, R. P. H.; Kanatzidis, M. G. Lead-Free Solid-State Organic-Inorganic Halide Perovskite Solar Cells. Nat. Photonics 2014, 8 (6), 489−494. (8) Noel, N. K.; Stranks, S. D.; Abate, A.; Wehrenfennig, C.; Guarnera, S.; Haghighirad, A.-A.; Sadhanala, A.; Eperon, G. E.; Pathak, S. K.; Johnston, M. B.; et al. Lead-Free Organic−Inorganic Tin Halide Perovskites for Photovoltaic Applications. Energy Environ. Sci. 2014, 7 (9), 3061−3068. (9) Ke, W.; Stoumpos, C. C.; Zhu, M.; Mao, L.; Spanopoulos, I.; Liu, J.; Kontsevoi, O. Y.; Chen, M.; Sarma, D.; Zhang, Y.; et al. Enhanced Photovoltaic Performance and Stability with a New Type of Hollow 3D Perovskite {en}FASnI3. Sci. Adv. 2017, 3 (8), e1701293. (10) Shao, S.; Liu, J.; Portale, G.; Fang, H.-H.; Blake, G. R.; ten Brink, G. H.; Koster, L. J. A.; Loi, M. A. Highly Reproducible Sn6030

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031

Letter

The Journal of Physical Chemistry Letters Hybrid Perovskite Engineered by Dual-Site Alloying. J. Phys. Chem. C 2017, 121 (17), 9508−9515. (28) Kim, J.; Lee, S.-H.; Lee, J. H.; Hong, K.-H. The Role of Intrinsic Defects in Methylammonium Lead Iodide Perovskite. J. Phys. Chem. Lett. 2014, 5 (8), 1312−1317. (29) Haruyama, J.; Sodeyama, K.; Han, L.; Tateyama, Y. Termination Dependence of Tetragonal CH3NH3PbI3 Surfaces for Perovskite Solar Cells. J. Phys. Chem. Lett. 2014, 5 (16), 2903−2909.

6031

DOI: 10.1021/acs.jpclett.8b02555 J. Phys. Chem. Lett. 2018, 9, 6024−6031