A Direct Bandgap Copper–Antimony Halide ... - ACS Publications

Communication pubs.acs.org/JACS

A Direct Bandgap Copper−Antimony Halide Perovskite Brenda Vargas,† Estrella Ramos,† Enrique Pérez-Gutiérrez,‡ Juan Carlos Alonso,† and Diego Solis-Ibarra*,† †

Instituto de Investigaciones en Materiales, Universidad Nacional Autónoma de México, CU, Coyoacán, 04510 Ciudad de México, México ‡ CONACYT-Laboratorio de Polímeros, Centro de Química, Instituto de Ciencias, Benemérita Universidad Autónoma de Puebla (BUAP), Complejo de Ciencias, ICUAP, C.P. 72570 Puebla, México S Supporting Information *

humidity. Our results expand the possible metal combinations and target structures to be used in the design of new perovskite materials with the overarching goal of replacing lead in PSCs with more benign elements. Numerous efforts have recently been devoted to replace lead in PSCs with other 2+ metals, but to date, only limited success has been achieved.3,7,8 With the pool of suitable 2+ metals almost exhausted, double perovskites, in which two metals (e.g., metals in 1+ and 3+ oxidation states) are combined to yield the same overall charge balance as conventional single-metal perovskites, have recently attracted significant attention.9−11 In this context, we decided to explore further the vast chemical space that the double perovskite architecture offers. We began our search by looking for alternative materials made of abundant and nontoxic elements. Given these criteria we decided to explore perovskites made of copper and antimony. Copper is by far the transition metal with the most reported halide perovskites (2D) and some of these materials have been used in solar cells, albeit with rather low PCEs (≤0.017%).12 On the other hand, we settled for antimony given its relative low toxicity,13 high abundance and reported ability to form extended structures made of corner-sharing SbX6 (X = Cl, Br, I) octahedra14,15 and even halide perovskites.16,17 Furthermore, SbIII is isoelectronic with SnII and can have good orbital overlap with copper, as demonstrated by the copper−antimony chalcogenides, that have been successfully implemented as absorbers in solar cells with PCEs of up to 3.5%.18 While double CuI−SbIII perovskites with the formula Cs2CuISbIIIX6 (X = Cl, Br and I) have been proposed and computationally examined,11 in our hands such compounds were inaccessible by traditional solution-state methods. However, although the combination of CuII−SbIII is not suited for a three-dimensional perovskite, it can still render a layered perovskite. Indeed, 1 can be obtained as a black microcrystalline powder when precipitated upon addition of stoichiometric amounts of CsCl to a solution of CuCl2 and Sb2O3 in hydrochloric acid (see Supporting Information for full experimental details). Single-crystalline square plates of up to 1 mm per side can be grown by slowly cooling down a solution of 1 in concentrated HCl; such crystals were used for single crystal X-ray diffraction (SCXRD) studies. SCXRD revealed a monoclinic C2/m system. The structure of 1 can be described

ABSTRACT: Since the establishment of perovskite solar cells (PSCs), there has been an intense search for alternative materials to replace lead and improve their stability toward moisture and light. As single-metal perovskite structures have yielded unsatisfactory performances, an alternative is the use of double perovskites that incorporate a combination of metals. To this day, only a handful of these compounds have been synthesized, but most of them have indirect bandgaps and/or do not have bandgaps energies well-suited for photovoltaic applications. Here we report the synthesis and characterization of a unique mixed metal ⟨111⟩-oriented layered perovskite, Cs4CuSb2Cl12 (1), that incorporates Cu2+ and Sb3+ into layers that are three octahedra thick (n = 3). In addition to being made of abundant and nontoxic elements, we show that this material behaves as a semiconductor with a direct bandgap of 1.0 eV and its conductivity is 1 order of magnitude greater than that of MAPbI 3 (MA = methylammonium). Furthermore, 1 has high photo- and thermal-stability and is tolerant to humidity. We conclude that 1 is a promising material for photovoltaic applications and represents a new type of layered perovskite structure that incorporates metals in 2+ and 3+ oxidation states, thus significantly widening the possible combinations of metals to replace lead in PSCs.

O

rganic−inorganic halide perovskites have emerged as one of the most promising families of photovoltaic materials. From the original report in 2009,1 power conversion efficiency (PCE) of solar cells based on these materials has risen from 3.8 to 22.1%.2 The pivotal material for this development has the formula MAPbI3, where MA stands for methylammonium. MAPbI3 and related materials have exhibit superb photophysical properties and amenable syntheses.3 However, the toxicity of lead and the lack of stability of these materials remain the main obstacles for big-scale implementation.4−6 Herein we present the synthesis and characterization of: Cs4CuIISb2IIICl12 (1, Figure 1A) which possesses a ⟨111⟩oriented triple-layered (n = 3, Figure 1A) perovskite structure and shows a direct bandgap of 1.0 eV. Besides being made of nontoxic and abundant elements, 1 can be conveniently synthesized in gram scale from aqueous solutions and shows no signs of decomposition upon exposure to light and/or © 2017 American Chemical Society

Received: April 24, 2017 Published: June 21, 2017 9116

DOI: 10.1021/jacs.7b04119 J. Am. Chem. Soc. 2017, 139, 9116−9119

Communication

Journal of the American Chemical Society

Figure 1. Family of ⟨111⟩-oriented perovskites with general formula: An+1BnX3n+3, can be obtained by cutting along the ⟨111⟩-direction of the 3D parent structure. Panels A and B show the crystal structures of Cs4CuSb2Cl12 (1) and α-Cs3Sb2Cl9 (2). Cl and Cs atoms are depicted as green and purple spheres, respectively; Sb and Cu coordination polyhedra are shown in gray and blue, respectively.

as a ⟨111⟩-oriented triple-layered (n = 3) perovskite of alternating, corner sharing CuCl6 and SbCl6 octahedra with cesium atoms occupying the voids in the framework (Figure 1A). The structure of 1 can also be visualized as a defective 3D perovskite, in which the 3+ charge on the antimony generates vacancies that result in the breakage of the 3D-network that yields a 2D network. The structure of 1 resembles an extended version of α-Cs3Sb2Cl9 (2, Figure 1a) where CuCl6 octahedra have been inserted between the SbCl6 layers.15 Both the CuCl6 and SbCl6 octahedra are significantly distorted (Figures S1 and S2). Most notably, the Cu−Cl distances are 2.299(1) and 2.808(1) Å for the equatorial and axial bonds, respectively. This disparity is characteristic of a Jahn−Teller distorted Cu2+ ion and it is additional (albeit indirect) confirmation of the oxidation state of the copper atom. The ⟨111⟩-oriented layered perovskite family with general formula An+1BnX3n+3 has several known members with n = 1 (zero-dimensional) and some others with n = 2. However, to the best of our knowledge, 1 is the first report of an n > 2 ⟨111⟩-oriented layered perovskite and it is also the first report of a mixed-metal halide layered perovskite. The lack of reports on these kind of structures can be understood from their general formula: An+1BnX3n+3, from which it can be inferred that for n values greater than two, the oxidation state on the B site must be fractional or a mixture of different B cations.16 This is in sharp contrast to other layered perovskite families, namely the ⟨100⟩- and ⟨110⟩-oriented perovskites, for which the oxidation state of B is preserved within a family regardless of the thickness of the inorganic layers.19,20 It is worth noting that lead-based ⟨100⟩-oriented perovskites have been successfully implemented as solar cells absorbers with PCEs of up to 12.5%.21,22 To assess the suitability of 1 for photovoltaic applications, we studied its optical absorption (Figure 2). The absorption spectrum of 1 signifies its semiconducting nature, while a Tauc plot assuming a direct transition, as taken from our DFT calculations (vide inf ra), results in a bandgap of 1.0 eV (Figure 2A). We note that the bandgap of 1 is well-suited for single absorber solar cells, with a theoretical maximum efficiency of 30.8% according to the Shockley−Queisser limit,23 or could also be used as the bottom absorber in an all-perovskite tandem solar cell. Notably, the bandgap of 1 is 2.0 eV smaller than that of its parent compound: 2 (3.0 eV, Figure S6). This drastic reduction of the bandgap is rather unusual for halide perovskites. For example, the bandgap reduction in lead halide

Figure 2. (A) Absorbance spectrum of 1. Inset: Tauc plot of 1 for a direct bandgap semiconductor. (B) Variable temperature conductivity (pressed pellets) and (C) Arrhenius plot of the temperature dependence of conductivity in 1. Vertical error bars in panel B correspond to instrumental uncertainty.

perovskites (Br, I or mixed halide) when going from a double (n = 2) to a triple layer (n = 3) structure is less than 0.3 eV.21,22,24,25 We also conducted variable temperature conductivity studies on pressed pellets of 1. The conductivity (σ) of 1 follows Arrhenius behavior, yielding a linear relationship between the ln(σ) and T−1 with and activation of 1.15 eV, a value that kindly matches its optical bandgap. Significantly, the conductivity of 1 is 1 order of magnitude higher than the one reported for pellets of MAPbI3,26 thus showing that despite its layered nature, the 9117

DOI: 10.1021/jacs.7b04119 J. Am. Chem. Soc. 2017, 139, 9116−9119

Communication

Journal of the American Chemical Society

tolerance toward humidity. Thermogravimetric analyses showed that 1 is stable up to 245 °C and differential scanning calorimetry indicates no phase transitions from −100 to +200 °C (Figure 4A,B). Exposure of 1 to relative humidities (RH) of

charge mobility in 1 is enough to be employed in a photovoltaic device. To gain a better insight into the properties of 1, we performed electronic structure calculations of 1 and 2 by density functional theory (DFT). Calculations were performed using the crystal structures of 1 and 227 and carried without further structure optimization. Band structure calculation for 1 resulted in a gap of 0.98 eV, in good agreement with our experimental results (Figure 3). The band structure of 1 shows

Figure 3. DFT-calculated band structure (left) and its projected density of states and partial density of states (DOS/pDOS, right) of 1. Occupied, partially occupied and unoccupied bands are displayed in gray, green and red, respectively.

a direct bandgap located near the E k point in the Brillouin zone. Significantly, the band structure of 1 is dispersive in both directions, perpendicular and parallel to the layers, meaning the layered nature of 1 may not handicap too much its use in photovoltaic devices.16,17 The density of states (DOS) and partial density of states (pDOS) calculations (Figure 3) show that the conduction band minimum (CBM) of 1 is predominantly antimony and chlorine p-orbital based. The latter is consistent with the behavior shown by 2 (Figure S8) and related material α-Cs3Sb2I917 for which the CBM is also antimony and halide p-based. The contribution of copper orbitals to the CBM is negligible and thus we can assume that the bandgap shrinkage is mainly due to the effect of copper in the valence band of 1. From the pDOS calculation (Figure 3), it can also be observed that the copper d-orbitals are located at the valence band maximum (VBM), which is also composed of chlorine p-orbitals. The large difference in the bandgap values of 1 and 2 can be explained by the good orbital overlap of the copper-d with the chlorine and antimony orbitals, which broadens the VBM and thus, reduces the gap. Notably, the presence of a d9 copper center also generates partially occupied bands at the top of the valence band, which may provide additional charge carriers. We also calculated the effective masses of holes and electrons in 1, the effective masses of holes and electrons are closely related to carrier mobility, which is a key factor on solar cell performance. The calculated effective masses for holes and electrons are 0.16 and 0.32 me respectively (me is the electron mass). These values are comparable to those calculated for MAPbI328 and suggest that 1 possesses high carrier mobility. Another important obstacle that perovskite materials face is the lack of stability toward heat, moisture and light. Therefore, we evaluated the photo and thermal stability of 1 as well as its

Figure 4. Stability of compound 1. Thermogravimetric analyses (A), differential scanning calorimetry (B) and PXRD of 1 after exposure to humidity and/or light (C).

50% showed no signs of decomposition for up to 100 days (Figures 4C and S10). Further, we evaluated the photostability of 1 under humid conditions. Remarkably, irradiation of 1 with a simulated sun under humid conditions (50% RH) did not decompose 1, nor did irradiation of 1 with UV-light for up to 15 days (Figures 4C and S12−S13). In conclusion, we have synthesized a new mixed-metal halide perovskite that incorporates two nontoxic and earth-abundant metals. Compound 1 has a direct bandgap and conductivity that makes it attractive for single-absorber and tandem solar cells. Further, 1 is highly stable to light, temperature and moisture and can be conveniently obtained by simple solutions methods in gram scales. Current efforts are being devoted to further understand its optoelectronic properties and to implement it in a solar cell. Simultaneously, efforts will be devoted to modulate the bandgap and optoelectronic properties of this material. In particular, doping or completely replacing copper with other first row transition metals presents a potentially viable and highly attractive alternative.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b04119. 9118

DOI: 10.1021/jacs.7b04119 J. Am. Chem. Soc. 2017, 139, 9116−9119

Communication

Journal of the American Chemical Society

■

(23) Rühle, S. Sol. Energy 2016, 130, 139. (24) Hu, H.; Salim, T.; Chen, B.; Lam, Y. M. Sci. Rep. 2016, 6, 33546. (25) Stoumpos, C. C.; Cao, D. H.; Clark, D. J.; Young, J.; Rondinelli, J. M.; Jang, J. I.; Hupp, J. T.; Kanatzidis, M. G. Chem. Mater. 2016, 28, 2852. (26) Yang, T. Y.; Gregori, G.; Pellet, N.; Grätzel, M.; Maier, J. Angew. Chem., Int. Ed. 2015, 54, 7905. (27) Kihara, K.; Sudo, T. Z. Kristallogr. - Cryst. Mater. 1971, 134, 142. (28) Filip, M. R.; Verdi, C.; Giustino, F. J. Phys. Chem. C 2015, 119, 25209−25219.

Detailed experimental procedures, crystallographic data, computational details (PDF) X-ray crystallographic file of 1 (CIF)

AUTHOR INFORMATION

Corresponding Author

*[email protected] ORCID

Enrique Pérez-Gutiérrez: 0000-0003-3761-4383 Diego Solis-Ibarra: 0000-0002-2486-0967 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank funding from PAPIIT IA203116/27 and CONACYT FC-2015-2/829. We also acknowledge the support of H. I. Karunadasa, M. Gembicky, A. Jaffe, M. Smith, M. Bizarro, A. Tejeda, E. Morales, A. Vivas, A. Pompa, M. Canseco and M. Olmos.

■

REFERENCES

(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am. Chem. Soc. 2009, 131, 6050. (2) NREL: National Center for Photovoltaics Home Page. http:// www.nrel.gov/pv/assets/images/efficiency_chart.jpg (accessed April 13, 2017). (3) Stoumpos, C. C.; Kanatzidis, M. G. Acc. Chem. Res. 2015, 48, 2791. (4) Flora, G.; Gupta, D.; Tiwari, A. Interdiscip. Toxicol. 2012, 5, 47. (5) Slavney, A. H.; Smaha, R. W.; Smith, I. C.; Jaffe, A.; Umeyama, D.; Karunadasa, H. I. Inorg. Chem. 2017, 56, 46. (6) Boix, P. P.; Agarwala, S.; Koh, T. M.; Mathews, N.; Mhaisalkar, S. G. J. Phys. Chem. Lett. 2015, 6, 898. (7) Hao, F.; Stoumpos, C. C.; Cao, D. H.; Chang, R.; Kanatzidis, M. G. Nat. Photonics 2014, 8, 489. (8) Ganose, A. M.; Savory, C. N.; Scanlon, D. O. Chem. Commun. 2017, 53, 20. (9) Slavney, A. H.; Hu, T.; Lindenberg, A. M.; Karunadasa, H. I. J. Am. Chem. Soc. 2016, 138, 2138. (10) McClure, E. T.; Ball, M. R.; Windl, W.; Woodward, P. M. Chem. Mater. 2016, 28, 1348. (11) Volonakis, G.; Filip, M. R.; Haghighirad, A. A.; Sakai, N.; Wenger, B.; Snaith, H. J.; Giustino, F. J. Phys. Chem. Lett. 2016, 7, 1254. (12) Cortecchia, D.; et al. Inorg. Chem. 2016, 55, 1044. (13) Smith, K. S.; Huyck, H. Rev. Econ. Geo. 1999, 6A, 29. (14) Ben Rhaiem, T.; Boughzala, H. Acta Crystallogr. E 2015, 71, 498. (15) Timmermans, C. W. M.; Cholakh, S. O.; Blasse, G. J. Solid State Chem. 1983, 46, 222. (16) Saparov, B.; Mitzi, D. B. Chem. Rev. 2016, 116, 4558. (17) Saparov, B.; Hong, F.; Sun, J.-P.; Duan, H.-S.; Meng, W.; Cameron, S.; Hill, I. G.; Yan, Y.; Mitzi, D. B. Chem. Mater. 2015, 27, 5622. (18) Welch, A. W.; Baranowski, L. L.; Zawadzki, P.; Lany, S.; Wolden, C. A.; Zakutayev, A. Appl. Phys. Express 2015, 8, 082301. (19) Mitzi, D.; Feild, C.; Harrison, W.; Guloy, A. Nature 1995, 369, 467. (20) Mitzi, D. B.; Wang, S.; Feild, C. A.; Chess, C. A.; Guloy, A. M. Science 1995, 267, 1473. (21) Smith, I. C.; Hoke, E. T.; Solis-Ibarra, D.; McGehee, M. D.; Karunadasa, H. I. Angew. Chem., Int. Ed. 2014, 53, 11232. (22) Tsai, H.; Nie, W.; Blancon, J.-C.; Stoumpos, C. C.; Asadpour, R.; Harutyunyan, B.; Neukirch, A. J.; Verduzco, R.; Crochet, J. J.; Tretiak, S.; Pedesseau, L.; Even, J.; Alam, M.; Gupta, G.; Lou, J.; Ajayan, P. M.; Bedzyk, M. J.; Kanatzidis, M. G.; Mohite, A. D. Nature 2016, 536, 312. 9119

DOI: 10.1021/jacs.7b04119 J. Am. Chem. Soc. 2017, 139, 9116−9119