Morphology and Carrier Extraction Study of Organic–Inorganic Metal

Letter pubs.acs.org/JPCL

Morphology and Carrier Extraction Study of Organic−Inorganic Metal Halide Perovskite by One- and Two-Photon Fluorescence Microscopy Xiaoming Wen,*,† Rui Sheng,† Anita W. Y. Ho-Baillie,† Aleš Benda,‡ Sanghun Woo,† Qingshan Ma,† Shujuan Huang,† and Martin A. Green† †

Australian Centre for Advanced Photovoltaics, University of New South Wales, Sydney 2052, Australia Biomedical Imaging Facility, Mark Wainwright Analytical Centre, University of New South Wales, Sydney 2052, Australia

‡

S Supporting Information *

ABSTRACT: The past two years have seen the uniquely rapid emergence of a new class of solar-cell-based on mixed organic−inorganic halide perovskite. In this work, we demonstrate a promising technique for studying the morphology of perovskite and its impact on carrier extraction by carrier transport layer using one-photon and two-photon fluorescence imaging in conjunction with time-resolved photoluminescence. This technique is not only effective in separating surface and bulk effects but it also allows the determination of lifetimes in localized regions and local carrier extraction efficiency. The difference in sensitivities of transport materials to grain boundaries and film uniformity is highlighted in this study. It is shown that the PCBM fabricated in this work is more sensitive to film nonuniformity, whereas spiro-OMeTAD is more sensitive to grain boundaries in terms of effective carrier extraction. SECTION: Energy Conversion and Storage; Energy and Charge Transport

T

laser scanning confocal microscope is capable of submicrometer axial resolution and, therefore, is suitable for three-dimensional optical sectioning,15 especially in conjunction with 2P excitation due to longer absorption depth. Under the 2P excitation, two coherent photons with energies less than the band gap of the material under investigation can be absorbed by the material if the excitation density is sufficient.16 The absorbed photons then excite electrons into the conduction band resulting in an emission of band edge photons.16,17 It should be noted that the process of two-photon absorption is sensitive to the excitation density related to the material’s nonlinear absorption coefficient.18,19 The 2P fluorescence imaging uses the infrared laser causing less damage to the sample.18,19 Confocal microscopy can be conveniently extended by fluorescence lifetime imaging (FLIM) modality allowing for microenvironment sensing studies through excited state lifetime monitoring and providing unique insight into the electron−hole dynamics in nanoscale region. Therefore, fluorescence imaging can further reveal regimes of electron−hole recombination providing insights into the carrier recombination mechanisms in the material or device. In this study, we report the use of one-photon and twophoton fluorescence microscopy in conjunction with lifetime measurements to investigate the morphology and its impact on

he past two years have seen the unprecedentedly rapid emergence of a new solar cell technology based on mixed organic−inorganic halide perovskite. It has been shown that the photovoltaic performance of these devices is greatly dependent on the film morphology, which itself depends on the deposition technique and subsequent treatment.1−5 Poor perovskite morphology is detrimental to device performance because it not only causes electrical shorting but also deleteriously impacts light harvesting, charge dissociation, transport, and recombination,3,6−8 particularly in planar junction devices.9−11 The large CH3NH3PbI3 grains and incomplete coverage resulting in pinholes have been observed in films prepared by nonoptimized conventional spin-coating methods,12,13 such as slow crystallization due to the high boiling point of DMF (N,Ndimethylformamide, 152 °C) and slow crystal growth due to slow nucleation rate during the drying process after spincoating. The morphology of perovskite is usually investigated by scanning electron microscopy (SEM) and transmission electron microscopy (TEM),1−4,13,14 which requires special specimen preparation and results in sample damage during scanning by the high energy incident beam. The inevitable exposure of the perovskite material to air and moisture makes it susceptible to decomposition. In contrast, fluorescence imaging via onephoton (1P) or two-photon (2P) excitation is contactless and can be done on films with a surface-protecting layer. It does not require special specimen preparation and does not cause film damage due to the absence of high-energy beam scanning.14 A © 2014 American Chemical Society

Received: September 22, 2014 Accepted: October 21, 2014 Published: October 21, 2014 3849

dx.doi.org/10.1021/jz502014r | J. Phys. Chem. Lett. 2014, 5, 3849−3853

The Journal of Physical Chemistry Letters

Letter

effective two-photon fluorescence as clear 2P imaging can also be obtained at modest excitation power. This suggests that perovskite has a large two-photon absorption coefficient suitable for application based on 2P absorption and fluorescence. The photoluminescence (PL) decay traces were measured using time correlated single photon counting (TCSPC) technique excited at 470 nm. Figure 2b shows the PL decays of perovskites before and after the addition of electron (PCBM) and hole (spiro-OMeTAD) transport layers in test structures as depicted in Figure 1. The perovskite exhibits nearly single exponential decay with a respectable lifetime of 79 ns (indicating sufficient quality perovskite film for achieving 12% conversion efficiency for an associated solar device, see Figure S1a in Supporting Information). When covered with the hole (spiro-OMeTAD) and electron (PCBM) transport materials, the lifetimes become significantly shorter, indicating effective carrier extraction or “quenching” by the transport materials, a key factor in high conversion efficiency in associated perovskite solar devices.22,25−27 It is interesting to note from biexponential fitting that in addition to the fast components, slower components with lifetimes of 109 and 93 ns are present for the perovskite/PCBM and perovskite/spiroOMeTAD structures, respectively. The long lifetime component suggests the existence of unquenched perovskites. The lifetimes (ca. 100 ns) of these unquenched perovskites are longer than that (ca. 80 ns) of the perovskite in the absence of transport layers because the carrier densities in the unquenched perovskites are lower resulting in lower carrier recombination rates. This is consistent with previous studies showing that the recombination rate is dependent on the carrier density in the perovskite.24−27 Figure 3 shows the one- and two-photon excited fluorescence images of the perovskite film before (a and b) and after (c and d) the addition of electron quenchers, detected with a 750/40 nm band-pass filter. High spatial resolution is demonstrated using the two-photon excited fluorescence imaging. It should be noted that under one-photon excitation (488 nm), the penetration depth is within the first 65 nm. Therefore, fluorescence originates from the vicinity of perovskite surface or quencher/perovskite interface with negligible contribution from the bulk. On the other hand, under two-photon excitation with longer wavelength absorbed deeper in the perovskite layer, bulk effects can be observed from the fluorescence response.

carrier extraction of organic−inorganic metal halide CH3NH3PbIXCl3−X perovskite by electron or hole transport layers; see Figure 1. In addition, the images obtained from 1P

Figure 1. Illustration of one- and two-photon excitation fluorescence imaging.

and 2P fluorescence imaging allow the separation of surface and bulk effects of carrier extraction and recombination providing useful information for device optimization. We found that grain boundaries and imperfect interface between perovskite and electron or hole transport layer are detrimental to carrier extraction but the severity varies depending on the type of transport layer used. It has been reported the CH3NH3PbIXCl3−X perovskite exhibits efficient fluorescence.20−22 Figure 2a shows the fluorescence spectrum of CH3NH3PbIXCl3−X perovskite excited at 405 nm and measured by a spectrometer, with a Si-CCD detector,23 as well as fluorescence spectra of the same material obtained by a Leica TCS SP5 microscope using 488 nm onephoton excitation and 950 nm two-photon excitation. The fluorescence spectra in the low energy side from the microscope were attenuated due to the low sensitivity of the detector in the near-infrared. The perovskite exhibits fluorescence maximum at 770 nm. This is consistent with other observations demonstrating CH3NH3PbIXCl3−X as a direct band gap material with fluorescence independent of the excitation wavelength. Therefore, perovskite based solar cells are suitable to be investigated by using the fluorescence microscope. By using a train of 950 nm femtosecond laser pulses, we observed a similar fluorescence spectrum under the 2P excitation to that measured under 1P excitation, indicating

Figure 2. (a) PL spectra of the CH3NH3PbIXCl3−X perovskite measured by a Si-CCD-spectrometer (one-photon, blue) with 405 nm excitation; and measured by a Leica TCS SP5 microscope with an excitation of 488 nm (one-photon, black) and 950 nm (two-photon, red). (b) PL decay traces of CH3NH3PbIXCl3−X perovskite (blue), perovskite/PCBM (black), and perovskite/spiro-OMeTAD (red) from a 50 × 50 μm2 region excited by 470 nm 5 MHz pulsed laser. 3850

dx.doi.org/10.1021/jz502014r | J. Phys. Chem. Lett. 2014, 5, 3849−3853

The Journal of Physical Chemistry Letters

Letter

Figure 4. (a) One-photon and (b) two-photon fluorescence images of epoxy/higher concentration spiro-OMeTD/perovskite/glass.

observed in these nonuniform regions (absence of significantly brighter fluorescence response under 2P imaging), interface recombination is adversely affected, characterized by the appearance of bright spots in Figure 3c in the nonuniform regions. The perovskite film nonuniformity can lead to PCBM film nonuniformity or lead to poor perovskite/PCBM interface. Either effect can result in carrier recombination before they are extracted by PCBM. On the other hand, spiro-OMeTAD covered perovskite exhibit different behavior whereby interface recombination is most significant along the grain boundary (see bright regions in Figure 4a) and less affected by film nonuniformity which only affects the bulk recombination (see Figure 4b). The dark regions within the grains in Figure 4a indicate efficient hole extraction suggesting good spiro-OMeTAD/perovskite interface can be generally achieved within the perovskite grain regardless of film uniformity. Apart from the ability to resolve surface and bulk effects, another advantage of the fluorescence imaging technique is the ability to carry out local lifetime measurements. Figure 5 shows

Figure 3. One-photon (1P; 488 nm excitation) and two-photon (2P; 950 nm excitation) fluorescence imaging with 750/40 nm band-pass filter, (a) 1P and (b) 2P images of perovskite/glass; (c) 1P and (d) 2P fluorescence images of PCBM/perovskite/glass.

Under both excitation regimes, there is negligible fluorescence contribution from the spiro-OMeTAD or the PCBM. The images show the comparable morphology characteristics of the perovskite film before and after the addition of electron or hole quenchers (see Figure S3 in Supporting Information) with the presence of perovskite grains as well as voids or pinholes (dark regions). The presence of pinholes in perovskite solar cells is undesirable and results in low conversion efficiency.3 Bright spots observed in the void or in the vicinity of the perovskite grain boundaries in the one-photon image, Figure 3a, suggest recombination activities. These bright spots are most likely isolated small grains from imperfect grain formation of perovskite. However, the bright spots were not observed in two-photon image suggesting these recombination activities are localized at the surface. Interference fringes are observed in the 2P images, Figure 3b,d due to interference between the fluorescence directly from the excitation point and the fluorescence reflected at the perovskite/glass interface.28 The bright outlines along the grain boundaries as observed in Figure 3b,d are due to the extra detected fluorescence reflected by the vertical face of the grain boundaries and then incident on the objective. These optical effects can be effectively removed after the addition of thick epoxy covering layer; see Figure 4b. 1P and 2P fluorescence images of the epoxy/spiroOMeTAD/perovkite/glass test structure are shown in Figure 4. Few interesting observations can be made by comparing images between Figure 3 and Figure 4. The first observation is indicated by the appearance of bright spots within the poor quality grain in Figure 3c. The poor quality grain in turn is a result of film nonuniformity (characterized by disruptions to the interference effects within the grains in Figure 3d), which is not uncommon as the film is fabricated using conventional spin coating and, therefore, susceptible to film nonuniformtiy.12 Although no significant increase in bulk recombination is

Figure 5. PL decay traces of “bright” and “dark” regions in CH3NH3PbIXCl3−X perovskite covered by PCBM and SpiroOMeTAD.

the PL decay traces of the dark and bright regions (0.5 × 0.5 μm2) of spiro-OMeTAD/perovskite and PCBM/perovskite test structures, which separates the binary components extracted from Figure 2b. The fast components in Figure 2b correspond to the dark regions of the fluorescence images as a result of efficient carrier extractions by the quenchers, whereas the slower components correspond to the brighter regions whereby unquenched carriers are left to recombine. We have demonstrated that one- and two-photon fluorescence microscopy is a promising technique for studying 3851

dx.doi.org/10.1021/jz502014r | J. Phys. Chem. Lett. 2014, 5, 3849−3853

The Journal of Physical Chemistry Letters

■

the morphology and its impact on carrier transport for perovskite cells. The two-photon images exhibit higher spatial definition with the extra advantage of depth definition, whereas one-photon images provide the information on the top surface or interface. Therefore, the combination of two imaging techniques is effective at separating the surface and bulk effects of carrier transport and recombination. This technique is also effective in separating components in PL decay traces that correspond to “good” and “bad” regions of the test or cell structure. This is most crucial for understanding the operation of perovskite solar cells and the optimization of the associated solar cell devices. It is shown that grain boundaries and imperfect interface between perovksite and electron or hole transport layer are detrimental to carrier extraction. It is shown that the PCBM fabricated in this work is more sensitive to film nonuniformity, whereas spiro-OMeTAD is more sensitive to grain boundaries in terms of effective carrier extraction.

■

Letter

ASSOCIATED CONTENT

S Supporting Information *

J−V curve and experimental methods of the fabrication and measurement of the associated solar device using the perovskite film fabricated in this work. XRD of perovskite film fabricated in this work. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The Australian Centre for Advanced Photonics (ACAP) encompasses the Australian-based activities of the Australia− U.S. Institute for Advanced Photovoltaics (AUSIAPV) and is supported by the Australian Government through the Australian Renewable Energy Agency (ARENA). Responsibility for the views, information, or advice expressed herein is not accepted by the Australian Government. We also thank the BioMedical Imaging Facility (BMIF), MWAC, UNSW for the fluorescence microscopy measurements and support.

EXPERIMENTAL METHODS

Methylammonium iodide (MAI) was synthesized following a previously reported method29 by reacting 24 mL of 0.20 mol methylamine (33 wt % in ethanol, Aldrich), 10 mL of 0.4 mol hydroiodic acid (57 wt % in water, Aldrich), and 100 mL ethanol in a 250 mL round-bottom flask at 0 °C for 2 h with stirring. After the reaction, the precipitate was recovered by using a rotary evaporator at 60 °C and then dissolved in ethanol followed by sedimentation in diethyl ether by stirring, which was repeated until the white MAI powder appears. The final product was collected and dried at 60 °C in an oven and dehydrated in a vacuum chamber. Perovskite precursor solution was prepared by dissolving PbCl2 and MAI in N,N-dimethylformamide (DMF) with a molar ratio of 1:1 under stirring at 70 °C. PbCl2 was purchased from Sigma-Aldrich and used as received. The CH3NH3PbIXCl3−X film was deposited by spin-coating the precursor solution at 2000 rpm for 60s on a borosilicate substrate, followed by annealing at 100 °C for 45 min. For the perovskite sample with an electron quencher, chlorobenzene solution of PC71BM with a concentration of 10 mg/mL was spin-coated onto the perovskite film at 2000 rpm for 60s. For the perovskite sample with a hole transport material, a solution was prepared by dissolving 72.3 mg (2,2′,7,7′-tetrakis(N,N-di-pmethoxyphenylamine)-9,9-spirobifluorene) (spiro-MeOTAD), 28.8 μL 4-tert-butylpyridine, 17.5 μL of a stock solution of 520 mg/mL lithium bis(trifluoromethylsulfonyl)imide in acetonitrile in 1 mL chlorobenzene. The hole transport material was deposited by spin-coating at 2000 rpm for 60s. To eliminate the interference fringes, a thick epoxy protective layer was deposited on top. One-photon and two-photon fluorescence imaging were carried out in a modified Leica TCS SP5 microscope. For onephoton imaging 488 nm line of continuous-wave argon-ion laser was used. For two-photon imaging, a mode-locked Ti:Sapphire laser (Mai Tai) producing 150 fs pulses at repetition rate of 80 MHz and tunable wavelength between 700 and 1060 nm was used. The PL decay traces were measured by microtime200 microscope (Picoquant) using TCSPC technique with excitation of 470 nm laser at 5 MHz repetition rate and detection through 750/40 nm band-pass filter. The experiment was undertaken at room temperature.

■

REFERENCES

(1) Nanova, D.; Kast, A. K.; Pfannmoeller, M.; Mueller, C.; Veith, L.; Wacker, I.; Agari, M.; Hermes, W.; Erk, P.; Kowalsky, W. Unraveling the Nanoscale Morphologies of Mesoporous Perovskite Solar Cells and Their Correlation to Device Performance. Nano Lett. 2014, 14, 2735−2740. (2) Kim, H.-B.; Choi, H.; Jeong, J.; Kim, S.; Walker, B.; Song, S.; Kim, J. Y. Mixed Solvents for the Optimization of Morphology in Solution-Processed, Inverted-Type Perovskite/Fullerene Hybrid Solar Cells. Nanoscale 2014, 6, 6679−6683. (3) Eperon, G. E.; Burlakov, V. M.; Docampo, P.; Goriely, A.; Snaith, H. J. Morphological Control for High Performance, SolutionProcessed Planar Heterojunction Perovskite Solar Cells. Adv. Funct. Mater. 2014, 24, 151−157. (4) Dualeh, A.; Tetreault, N.; Moehl, T.; Gao, P.; Nazeeruddin, M. K.; Graetzel, M. Effect of Annealing Temperature on Film Morphology of Organic−Inorganic Hybrid Pervoskite Solid-State Solar Cells. Adv. Funct. Mater. 2014, 24, 3250−3258. (5) Green, M. A.; Ho-Baillie, A.; Snaith, H. J. The Emergence of Perovskite Solar Cells. Nat. Photo. 2014, 8, 506−514. (6) Jeng, J.-Y.; Chiang, Y.-F.; Lee, M.-H.; Peng, S.-R.; Guo, T.-F.; Chen, P.; Wen, T.-C. CH3NH3PbI3 Perovskite/Fullerene PlanarHeterojunction Hybrid Solar Cells. Adv. Mater. 2013, 25, 3727−3732. (7) Docampo, P.; Ball, J. M.; Darwich, M.; Eperon, G. E.; Snaith, H. J. Efficient Organometal Trihalide Perovskite Planar-Heterojunction Solar Cells on Flexible Polymer Substrates. Nat. Commun. 2013, 4, 2761. (8) Conings, B.; Baeten, L.; De Dobbelaere, C.; D’Haen, J.; Manca, J.; Boyen, H.-G. Perovskite-Based Hybrid Solar Cells Exceeding 10% Efficiency with High Reproducibility Using a Thin Film Sandwich Approach. Adv. Mater. 2014, 26, 2041−2046. (9) Yang, X.; Loos, J. Toward High-Performance Polymer Solar Cells: The Importance of Morphology Control. Macromolecules 2007, 40, 1353−1362. (10) Wang, Z.-S.; Kawauchi, H.; Kashima, T.; Arakawa, H. Significant Influence of TiO2 Photoelectrode Morphology on the Energy Conversion Efficiency of N719 Dye-Sensitized Solar Cell. Coord. Chem. Rev. 2004, 248, 1381−1389. (11) Jaegermann, W.; Klein, A.; Mayer, T. Interface Engineering of Inorganic Thin-Film Solar Cells − Materials-Science Challenges for Advanced Physical Concepts. Adv. Mater. 2009, 21, 4196−4206. 3852

dx.doi.org/10.1021/jz502014r | J. Phys. Chem. Lett. 2014, 5, 3849−3853

The Journal of Physical Chemistry Letters

Letter

(12) Burschka, J.; Pellet, N.; Moon, S.-J.; Humphry-Baker, R.; Gao, P.; Nazeeruddin, M. K.; Graetzel, M. Sequential Deposition as a Route to High-Performance Perovskite-Sensitized Solar Cells. Nature 2013, 499, 316. (13) Xiao, M.; Huang, F.; Huang, W.; Dkhissi, Y.; Zhu, Y.; Etheridge, J.; Gray-Weale, A.; Bach, U.; Cheng, Y. B.; Spiccia, L. A Fast Deposition-Crystallization Procedure for Highly Efficient Lead Iodide Perovskite Thin-Film Solar Cells. Angew. Chem. 2014, 126, 10056− 10061. (14) Hao, X.-T.; McKimmie, L. J.; Smith, T. A. Spatial Fluorescence Inhomogeneities in Light-Emitting Conjugated Polymer Films. J. Phys. Chem. Lett. 2011, 2, 1520−1525. (15) Wilson, T. Confocal Microscopy; Academic Press: London, 1990; Vol. 426, pp 1−64. (16) Wen, X.; Yu, P.; Toh, Y.-R.; Ma, X.; Tang, J. On the Upconversion Fluorescence in Carbon Nanodots and Graphene Quantum Dots. Chem. Commun. 2014, 50, 4703−4706. (17) Im, J.-H.; Lee, C.-R.; Lee, J.-W.; Park, S.-W.; Park, N.-G. 6.5% Efficient Perovskite Quantum-Dot-Sensitized Solar Cell. Nanoscale 2011, 3, 4088−4093. (18) Wen, X.; Xu, P.; Lukins, P. B.; Ohno, N. Confocal Two-Photon Spectroscopy of Red Mercuric Iodide. Appl. Phys. Lett. 2003, 83, 425− 427. (19) Denk, W.; Strickler, J. H.; Webb, W. W. Two-Photon Laser Scanning Fluorescence Microscopy. Science 1990, 248, 73−76. (20) Deschler, F.; Price, M.; Pathak, S.; Klintberg, L. E.; Jarausch, D.D.; Higler, R.; Huettner, S.; Leijtens, T.; Stranks, S. D.; Snaith, H. J. High Photoluminescence Efficiency and Optically Pumped Lasing in Solution-Processed Mixed Halide Perovskite Semiconductors. J. Phys. Chem. Lett. 2014, 5, 1421−1426. (21) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Graetzel, M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- and Hole-Transport Lengths in Organic−Inorganic CH3NH3PbI3. Science 2013, 342, 344−347. (22) Sum, T. C.; Mathews, N. Advancements in Perovskite Solar Cells: Photophysics Behind the Photovoltaics. Eng. Environ. Sci. 2014, 7, 2518−2534. (23) Wang, F.; Jiang, Y.; Wen, X.; Xia, J.; Sha, G.; Amal, R. Confined Au−Pd Ensembles in Mesoporous TiO2 Spheres for the Photocatalytic Oxidation of Acetaldehyde. ChemCatChem 2013, 5, 3557−3561. (24) Zhao, Y.; Zhu, K. CH3NH3Cl-Assisted One-Step Solution Growth of CH3NH3PbI3: Structure, Charge-Carrier Dynamics, and Photovoltaic Properties of Perovskite Solar Cells. J. Phys. Chem. C 2014, 118, 9412−9418. (25) Wehrenfennig, C.; Liu, M.; Snaith, H. J.; Johnston, M. B.; Herz, L. M. Charge-Carrier Dynamics in Vapour-Deposited Films of the Organolead Halide Perovskite CH3NH3PbI3−xClx. Eng. Environ. Sci. 2014, 7, 2269−2275. (26) Roiati, V.; Colella, S.; Lerario, G.; De Marco, L.; Rizzo, A.; Listorti, A.; Gigli, G. Investigating Charge Dynamics in Halide Perovskite-Sensitized Mesostructured Solar Cells. Eng. Environ. Sci. 2014, 7, 1889−1894. (27) Yamada, Y.; Nakamura, T.; Endo, M.; Wakamiya, A.; Kanemitsu, Y. Photocarrier Recombination Dynamics in Perovskite CH3NH3PbI3 for Solar Cell Applications. J. Am. Chem. Soc. 2014, 136, 11610− 11613. (28) Wen, X.; Smith, T.; Ghiggino, K.; Dao, L.; Hannaford, P. Observation of Back-Surface Reflected Luminescence in GaAs Excited by Ultrashort Pulses. Appl. Phys. Lett. 2009, 94, 102101. (29) Liang, P.-W.; Liao, C.-Y.; Chueh, C.-C.; Zuo, F.; Williams, S. T.; Xin, X.-K.; Lin, J.; Jen, A. K. Y. Additive Enhanced Crystallization of Solution-Processed Perovskite for Highly Efficient Planar-Heterojunction Solar Cells. Adv. Mater. 2014, 26, 3748−3754.

3853

dx.doi.org/10.1021/jz502014r | J. Phys. Chem. Lett. 2014, 5, 3849−3853