Phase Transitions of Formamidinium Lead Iodide Perovskite under

Article Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

pubs.acs.org/JACS

Phase Transitions of Formamidinium Lead Iodide Perovskite under Pressure Shaojie Jiang,† Yiliang Luan,‡ Joon I. Jang,§ Tom Baikie,*,∥ Xin Huang,⊥ Ruipeng Li,⊥,⊗ Felix O. Saouma,# Zhongwu Wang,*,⊥ Timothy J. White,¶ and Jiye Fang*,†,‡

Downloaded via UNIV OF LOUISIANA AT LAFAYETTE on October 13, 2018 at 11:28:17 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Materials Science and Engineering Program, State University of New York at Binghamton, Binghamton, New York 13902, United States ‡ Department of Chemistry, State University of New York at Binghamton, Binghamton, New York 13902, United States § Department of Physics, Sogang University, 35 Baekbeom-ro, Mapo-gu, Seoul 04107, South Korea ∥ Energy Research Institute@NTU, Nanyang Technological University, Singapore 639798, Republic of Singapore ⊥ Cornell High Energy Synchrotron Source (CHESS), Cornell University, Ithaca, New York 14853, United States # Department of Physical Sciences, Kaimosi Friends University College, P.O. Box 385-50309, Kaimosi, Kenya ¶ School of Materials Science and Engineering, Nanyang Technological University, Singapore 639798, Republic of Singapore S Supporting Information *

ABSTRACT: The pressure-induced structural evolution of formamidinium-based perovskite FAPbI3 was investigated using in situ synchrotron X-ray diffraction and laserexcited photoluminescence methods. Cubic α-FAPbI3 (Pm3̅ m) partially and irreversibly transformed to hexagonal δ-FAPbI3 (P63mc) at a pressure less than 0.1 GPa. Structural transitions of α-FAPbI3 followed the sequence of Pm3̅m → P4/mbm → Im3̅ → partial amorphous during compression to 6.59 GPa, whereas the δ-phase converted to an orthorhombic Cmc21 structure between 1.26 and 1.73 GPa. During decompression, FAPbI3 recovered the P63mc structure of the δ-phase as a minor component (∼18 wt %) from 2.41−1.40 GPa and the Pm3̅m structure of the α-phase becomes dominant (∼82 wt %) at 0.10 GPa but with an increased fraction of δ-FAPbI3. The photoluminescence behaviors from both the α- and δ-forms were likely controlled by radiative recombination at the defect levels rather than band-edge emission during pressure cycling. FAPbI3 polymorphism is exquisitely sensitive to pressure. While modest pressures can engineer FAPbI3-based photovoltaic devices, irreversible δ-phase crystallization may be a limiting factor and should be taken into account.

■

INTRODUCTION Hybrid perovskite-based photovoltaic (PV) cells have shown remarkable improvements in power conversion efficiency (PCE), increasing from ∼3.8% in 20091 to more than 21% in 2017.2 Sunlight possesses a panchromatic spectrum, requiring absorber materials with a certain bandgap. It is known that the highest PCE at 300 K for 1 sun correlates with a bandgap of 1.2−1.4 eV.3,4 MAPbX3 (MA: methylammonium; X: halide anions) is the most widely studied hybrid perovskite with the smallest bandgap5−8 of 1.57 eV which is larger than the optimal value. The bandgap can be reduced by introducing organic groups with a larger effective ionic radius (R). For example, a replacement of MA (R = 2.70 Å)9 with a slightly larger formamidinium cation (NH2CHNH2+, FA, R = 2.79 Å)9 in the perovskite MAPbI3 can dramatically reduce the bandgap to ∼1.45 eV,5 making FAPbI3 a more suitable absorber for single-junction solar cells.10 It is also reported that FAPbI3 has a high short-circuit photocurrent without sacrificing the photovoltage owing to more efficient harvesting of red photons.11 Furthermore, FAPbI3 exhibits superior © XXXX American Chemical Society

charge transport parameters including carrier mobility compared to MAPbI3.12 Therefore, exploration of FAPbI3 polymorphism and the relationship to PV functionality is required. FAPbI3 polymorphism as a function of temperature has been extensively studied. At room temperature, FAPbI3 crystallizes as a yellow, nonperovskite δ-phase (P63mc) which is thermally stable. However, this material is a poor solar absorbent with a large bandgap and a chain-like structure that hinders electron transport.10 At ∼150 °C, the δ-phase FAPbI3 (hereafter, δFAPbI3) undergoes a reconstructive phase transformation to the α-phase FAPbI3 (hereafter, α-FAPbI3) with a favorable bandgap and topology for efficient PV effects.13 The α-FAPbI3 is a cubic perovskite (Pm3̅m) where [PbI6]4− octahedra are vertex-connected.14 At room temperature, the α-FAPbI3 slowly transforms to the thermodynamically stable δ-FAPbI 3 irrespective of the storage environment.13 Recent studies Received: August 29, 2018 Published: September 28, 2018 A

DOI: 10.1021/jacs.8b09316 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society have focused on α-FAPbI3 stability15−17 and its application in PV cells,4,18,19 whereas pressure-dependent investigations on FAPbI3 are very limited.20 The chemical analog FAPbBr3 transforms from Pm3̅m to Im3̅ as the pressure increased from ambient pressure to 0.53 GPa, leading to a bandgap redshift followed by a blue-shift at a higher pressure (2.2 GPa).21 Similarly, MAPbI3 evolves from I4/mcm to Im3̅ below 0.3 GPa with a bandgap red-shift followed by a blue-shift upon further compression.22 This phenomenon has not been reported in FAPbI3.20 Thus, the motivation for the present high-pressure experiments using synchrotron X-ray diffraction (XRD) and photoluminescence (PL) techniques was to clarify the relationship between the crystallographic and electronic structures of the FAPbI3 polymorphs and the implications for photovoltaic deployment of this material.

■

series diode-pumped picosecond Nd:YAG laser with a pulse width of 30 ps and a repetition rate of 50 Hz. The Nd:YAG laser pumped an EKSPLA Harmonics Unit (HU) H400 where the input beam was frequency tripled by a successive cascade of nonlinear wave mixing to obtain the UV excitation source. Fitting Software. Pawley fitting of the lattice parameters and Rietveld refinements for quantitative analysis of phase component were carried out using Topas (version 3) software package (1999− 2000 Bruker AXS). Previously reported lattice parameters of αFAPbI3 at ambient pressure as determined by neutron diffraction14 were used as the starting point for refinement. The corresponding synchrotron XRD pattern was simulated using CrystalMaker.24

■

RESULTS AND DISCUSSION Freshly annealed black FAPbI3 at ambient pressure was found by synchrotron powder diffraction to be phase pure α-FAPbI3 (Figure 1a,b). Pawley fitting (Figure S1a) verified Pm3̅m symmetry14 with a cell edge of a = 6.3651(1) Å.

EXPERIMENTAL SECTION

Sample Preparation. FA- and lead-containing solutions were separately prepared by dissolving 1.9 mmol of formamidine acetate (99%, Sigma-Aldrich) in 0.5 mL of hydriodic acid (HI, 57 wt % in water, Sigma-Aldrich) and dissolving 1.8 mmol lead(II) acetate trihydrate (≥99%, Sigma-Aldrich) in 4.0 mL of HI in a flask heated in an oil bath (100 °C). The FA-containing solution was then slowly introduced to the Pb-containing solution in a round-bottom flask, generating a yellow precipitate. Additional HI solution was added dropwise to this mixture at 100 °C under stirring until the yellow precipitate was completely dissolved. This solution was maintained at 100 °C for 2 h under vigorous stirring, then slowly cooled to 75 °C and held at this temperature overnight without stirring to yield yellow acicular FAPbI3 crystals. The product was harvested by filtration, washed using diethyl ether (J & T Baker), and dried in a vacuum oven at room temperature. Black α-FAPbI3 (Pm3̅ m) powders were obtained by heating the yellow crystals at 170 °C for 30 min in argon. A portion of the black FAPbI3 product was loaded to a diamond anvil cell (DAC), whereas the unused material was sealed in a vial at ambient pressure as a “reference” specimen for future phase comparison. High-Pressure Setup. The DAC setup was prepared by preindenting the stainless steel gasket to a thickness of ∼100 μm from 250 μm through which a ∼200 μm diameter hole was drilled and served as the sample chamber. The perovskite was loaded with several ruby chips to calibrate the pressure by laser-excited ruby fluorescence technique during the in situ experiments. Synchrotron XRD patterns and PL spectra were collected successively. X-ray Diffraction and Photoluminescence Characterizations. Synchrotron powder XRD patterns were collected at the B1 station of the Cornell High Energy Synchrotron Source (CHESS). Monochromatic X-rays (25.514 keV) was collimated using a double pinhole-aligned circular tube into the small X-ray beam (100 μm in diameter). X-ray scattering signals were collected using a large area Mar345 detector. The sample-to-detector distance and other detector parameters were calibrated using a CeO2 powder standard. The raw two-dimensional (2D) images were integrated and analyzed by the Fit2D package.23 The laser-excited PL spectra were recorded using a Princeton Acton SP-300i system from 660 to 940 nm. The samples loaded in the DAC were measured under compression or decompression with in situ measurements of each diffraction pattern. A 532 nm diode laser was used for excitation and the emitted light was collected through a spectrometer with a 300g/mm grating and exposure time of 3s. In order to make a consistent comparison and correlation between the collected diffraction and PL data sets, the pressure-dependent compression and decompression experiments were repeated in the same manner. A time-integrated photoluminescence (TIPL) spectrum of the δ-FAPbI3 was also measured at ambient conditions under UV radiation. The sample was flamesealed in a quartz capillary tube, and loaded into a homemade sample holder that was mounted on a Z-scan translation stage. Coherent light (wavelength: 1064 nm) was first produced using an EKSPLA PL-2250

Figure 1. XRD patterns of α-FAPbI3 powder. (a) Experimental pattern at ambient pressure; (b) calculated pattern at ambient pressure; (c) experimental pattern at 0.10 GPa, showing the minor peaks (with stars) of δ-FAPbI3. The inset in panel c shows the structure of δ-FAPbI3.

For compression at 2.88 GPa, the feature was no longer detected (Figure 4 and Table 1, Section II). Meanwhile, a new PL peak emerges at 854 nm attributable to the P4/mbm polymorph from 0.46 GPa that red-shifts to ∼868 nm at 2.88

Figure 4. PL spectra of α-FAPbI3 upon increase of pressure.

Table 1. PL Peaks of α-FAPbI3 as a Function of Pressure Applied during Compression Section I

Section II

Section III

Pressure (GPa)

λ1 (nm)

E1 (eV)

λ2 (nm)

E2 (eV)

λ3 (nm)

E3 (eV)

0.10 0.24 0.46 0.93 1.26 1.51 1.97 2.30 2.55 2.88 3.13

724.9 725.4 726.9 − − − − − − 732.1 −

1.711 1.709 1.706 − − − − − − 1.694 −

795.8 796.7 800.3 804.8 805.8 807.3 809.3 809.8 808.3 803.6 799.3

1.558 1.556 1.549 1.541 1.539 1.536 1.532 1.531 1.534 1.543 1.551

853.1 853.5 854.8 855.7 857.1 859.6 864.3 867.5 867.7 867.9 868.1

1.454 1.453 1.451 1.449 1.447 1.443 1.435 1.429 1.429 1.429 1.428

GPa (Figure 4 and Table 1, Section III). After transformation to the Im3̅ polymorph at 3.13 GPa and at higher pressures the PL peaks became too weak for analysis (Figure 4). These observations suggest that the bandgap is strongly affected by the pressure-induced reorientation of the [PbI6]4− octahedra and possibly other structural factors. While the red-shift of these peaks may be indicative of a narrower bandgap, this could not be confirmed. A plausible explanation is that all of them originate from the defect levels common to the material, regardless of a specific phase. Consequently, they are not directly associated with the bandgap and ineffective for light harvesting. During decompression, both the PL peaks are restored from 2.41 and 1.4 GPa when the Im3̅ and Cmc21 crystallizes (Figure 5 and Table 2, Sections I and II). The PL peak at ∼718 nm corresponds to the transition from the Cmc21 to the P63mc structure in δ-FAPbI3 that comprises a larger fraction at this D

DOI: 10.1021/jacs.8b09316 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society

δ-FAPbI3 P63mc by compression to 0.1 GPa. The remaining sample fraction undergoes displacive transformation to three FAPbI3 polymorphs in the sequence of Pm3̅m → P4/mbm → Im3̅ → partially amorphous from ambient pressure to 6.59 GPa (Figure 6, left). Concomitantly, δ-FAPbI3 adopts a Cmc21

Figure 6. Illustrations of the phase transition of α-FAPbI3 during compression and decompression.

structure from 1.26 to 1.73 GPa. Upon decompression, the αFAPbI3 and δ-FAPbI3 dimorphs are restored (Figure 6, right) in the proportions ∼82 and ∼18 wt %, respectively. The PL behavior for α- and δ-FAPbI3 are most likely associated with transitions involving the defect levels rather than the bandgap variations only. Upon compression, a slight reduction of the αFAPbI3 bandgap before 1.51 GPa is indicated but further confirmation using techniques such as absorption spectroscopy31,32 is necessary. It is concluded that pressure induces both displacive and reconstructive transitions in FAPbI3 polymorphs including irreversible conversion to δ-FAPbI3. In order to promote the photon harvest efficiency of α-FAPbI3 through modification of its structure and bandgap by pressurization, the accelerated phase transition to the hexagonal nonperovskite structure (δ-FAPbI3) should be considered.

Figure 5. PL spectra of α-FAPbI3 upon release of pressure.

Table 2. PL Peaks of α-FAPbI3 as a Function of Pressure Applied during Decompression Section I

Section II

Section III

Pressure (GPa)

λ1 (nm)

E1 (eV)

λ2 (nm)

E2 (eV)

λ3 (nm)

E3 (eV)

1.40 1.15 0.46 0.10

718.1 717.8 724.7 726.4

1.727 1.728 1.711 1.707

784.4 784.3 799.7 776.7

1.581 1.581 1.551 1.597

− − 856.5 −

− − 1.448 −

■

pressure, whereas the peak at ∼784 nm is assigned to the transition from the Im3̅ to the P4/mbm polymorph in αFAPbI3. The PL peak at ∼860 nm was not detected during decompression, suggesting it is not associated with the bandgap but reflects a decrease in defect concentration during recrystallization. The peak position and intensity remain steady as the pressure was further decreased to 1.15 GPa (Figure 5), but at 0.46 GPa, the 718 nm feature weakens and red-shifts to 724 nm, whereas the second peak at 784 nm intensifies and red-shifts to 800 nm. An additional weak peak appears at 856 nm at this pressure (Figure 5 and Table 2, Section III) but is lost at 0.10 GPa when two additional peaks become evident at 726 and 776 nm. The differences in PL features during compression/decompression can be attributed to varying fractions of δ-FAPbI3. Concerning the lattice cell change, Figure S10 and Table S3 show a variation of the unit cell volume as a function of applied pressure during compression. This comparison indicates that the cell volume contractions at pressures of 0.24 and 2.88 GPa, which are associated with the α-FAPbI3 phase transitions of Pm3̅m → P4/mbm and P4/mbm → Im3̅, are ∼2.2% and ∼1.4%, respectively.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.8b09316. Pawley fitting of α- and δ-FAPbI3 during compression and decompression, discussion on diffraction and PL spectra of δ-FAPbI3 under pressure, XRD patterns of δFAPbI3 under pressure, PL spectra and peak positions of δ-FAPbI3 at ambient and high pressures, selected profiles of Rietveld fitting, change of lattice cell volume as a function of applied pressure (PDF)

■

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] *[email protected] ORCID

Joon I. Jang: 0000-0002-1608-8321 Zhongwu Wang: 0000-0001-9742-5213 Jiye Fang: 0000-0003-3703-3204

■

CONCLUSIONS An in situ XRD and PL investigation found α-FAPbI3 Pm3̅m perovskite to partially and irreversibly convert to nonperovskite

Present Address

⊗ National Synchrotron Light Source II, Brookhaven National Laboratory, Upton, New York 11973, USA

E

DOI: 10.1021/jacs.8b09316 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Article

Journal of the American Chemical Society Notes

(24) Palmer, D. C. Z. Krist. - Cryst. Mater. 2015, 230, 559−572. (25) Chen, T.; Foley, B. J.; Park, C.; Brown, C. M.; Harriger, L. W.; Lee, J.; Ruff, J.; Yoon, M.; Choi, J. J.; Lee, S. H. Sci. Adv. 2016, 2, e1601650. (26) Zheng, X. J.; Wu, C. C.; Jha, S. K.; Li, Z.; Zhu, K.; Priya, S. ACS Energy Lett. 2016, 1, 1014−1020. (27) Glazer, A. M. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1975, 31, 756−762. (28) Howard, C. J.; Stokes, H. T. Acta Crystallogr., Sect. B: Struct. Sci. 1998, 54, 782−789. (29) Young, J.; Rondinelli, J. M. J. Phys. Chem. Lett. 2016, 7, 918− 922. (30) Szafranski, M.; Katrusiak, A. J. Phys. Chem. Lett. 2016, 7, 3458− 3466. (31) Xiao, G.; Cao, Y.; Qi, G.; Wang, L.; Liu, C.; Ma, Z.; Yang, X.; Sui, Y.; Zheng, W.; Zou, B. J. Am. Chem. Soc. 2017, 139, 10087− 10094. (32) Zhang, L.; Liu, C.; Wang, L.; Liu, C.; Wang, K.; Zou, B. Angew. Chem., Int. Ed. 2018, 57, 11213−11217.

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge Custom Electronics Inc. for the financial support. We also thank Kevin K. Fang for his contribution to the experimental determination. This work is based upon research conducted at the Cornell High Energy Synchrotron Source (CHESS) which is supported by the National Science Foundation under award DMR-1332208. S. J. and Y.L. acknowledge the partial support from ACS PRF (PRF# 58196-ND10). J.I.J. acknowledges the support 2017R1D1A1B03035539 from NRF of Korea.

■

REFERENCES

(1) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. J. Am. Chem. Soc. 2009, 131, 6050−6051. (2) Yang, W. S.; Park, B.-W.; Jung, E. H.; Jeon, N. J.; Kim, Y. C.; Lee, D. U.; Shin, S. S.; Seo, J.; Kim, E. K.; Noh, J. H.; Seok, S. I. Science 2017, 356, 1376−1379. (3) Hao, F.; Stoumpos, C. C.; Chang, R. P. H.; Kanatzidis, M. G. J. Am. Chem. Soc. 2014, 136, 8094−8099. (4) Jeon, N. J.; Noh, J. H.; Yang, W. S.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. Nature 2015, 517, 476−481. (5) Eperon, G. E.; Stranks, S. D.; Menelaou, C.; Johnston, M. B.; Herz, L. M.; Snaith, H. J. Energy Environ. Sci. 2014, 7, 982−988. (6) Noh, J. H.; Im, S. H.; Heo, J. H.; Mandal, T. N.; Seok, S. I. Nano Lett. 2013, 13, 1764−1769. (7) Zhao, Y.; Zhu, K. J. Am. Chem. Soc. 2014, 136, 12241−12244. (8) Wang, B.; Xiao, X.; Chen, T. Nanoscale 2014, 6, 12287−12297. (9) Amat, A.; Mosconi, E.; Ronca, E.; Quarti, C.; Umari, P.; Nazeeruddin, M. K.; Gratzel, M.; De Angelis, F. Nano Lett. 2014, 14, 3608−3616. (10) Koh, T. M.; Fu, K. W.; Fang, Y. N.; Chen, S.; Sum, T. C.; Mathews, N.; Mhaisalkar, S. G.; Boix, P. P.; Baikie, T. J. Phys. Chem. C 2014, 118, 16458−16462. (11) Pellet, N.; Gao, P.; Gregori, G.; Yang, T. Y.; Nazeeruddin, M. K.; Maier, J.; Gratzel, M. Angew. Chem., Int. Ed. 2014, 53, 3151−3157. (12) Zhumekenov, A. A.; Saidaminov, M. I.; Haque, M. A.; Alarousu, E.; Sarmah, S. P.; Murali, B.; Dursun, I.; Miao, X. H.; Abdelhady, A. L.; Wu, T.; Mohammed, O. F.; Bakr, O. M. ACS Energy Lett. 2016, 1, 32−37. (13) Han, Q. F.; Bae, S. H.; Sun, P. Y.; Hsieh, Y. T.; Yang, Y.; Rim, Y. S.; Zhao, H. X.; Chen, Q.; Shi, W. Z.; Li, G.; Yang, Y. Adv. Mater. 2016, 28, 2253−2258. (14) Weller, M. T.; Weber, O. J.; Frost, J. M.; Walsh, A. J. Phys. Chem. Lett. 2015, 6, 3209−3212. (15) Ma, F. S.; Li, J. W.; Li, W. Z.; Lin, N.; Wang, L. D.; Qiao, J. Chem. Sci. 2017, 8, 800−805. (16) Binek, A.; Hanusch, F. C.; Docampo, P.; Bein, T. J. Phys. Chem. Lett. 2015, 6, 1249−1253. (17) Lee, J. W.; Kim, D. H.; Kim, H. S.; Seo, S. W.; Cho, S. M.; Park, N. G. Adv. Energy Mater. 2015, 5, 1501310. (18) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. Science 2015, 348, 1234−1237. (19) Wu, C. C.; Zheng, X. J.; Yang, Q.; Yan, Y. K.; Sanghadasa, M.; Priya, S. J. Phys. Chem. C 2016, 120, 26710−26719. (20) Wang, P.; Guan, J.; Galeschuk, D. T. K.; Yao, Y.; He, C. F.; Jiang, S.; Zhang, S.; Liu, Y.; Jin, M.; Jin, C.; Song, Y. J. Phys. Chem. Lett. 2017, 8, 2119−2125. (21) Wang, L. R.; Wang, K.; Zou, B. J. Phys. Chem. Lett. 2016, 7, 2556−2562. (22) Jiang, S.; Fang, Y.; Li, R.; Xiao, H.; Crowley, J.; Wang, C.; White, T. J.; Goddard, W. A., III; Wang, Z.; Baikie, T.; Fang, J. Angew. Chem., Int. Ed. 2016, 55, 6540−6544. (23) Hammersley, A. P. FIT2D: An Introduction and Overview (ESRF Internal Report); ESRF: 1997; p ESRF97HA02T. F

DOI: 10.1021/jacs.8b09316 J. Am. Chem. Soc. XXXX, XXX, XXX−XXX