Transient Sub-bandgap States in Halide Perovskite Thin Films - Nano

3 days ago - Note, however, that the initial carrier temperature and cooling rates are not uniform, a consequence of the large degree of heterogeneity...
0 downloads 20 Views 1MB Size
Subscriber access provided by Stony Brook University | University Libraries

Communication

Transient Sub-bandgap States in Halide Perovskite Thin Films Sanghee Nah, Boris Spokoyny, Xinyi Jiang, Costas Stoumpos, Chan Soe, Mercouri G. Kanatzidis, and Elad Harel Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.7b04078 • Publication Date (Web): 30 Jan 2018 Downloaded from http://pubs.acs.org on January 31, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Transient Sub-bandgap States in Halide Perovskite Thin Films S. Nah1, B. Spokoyny1, X. Jiang, C. Stoumpos1, C.M.M Soe1, 2, M.G. Kanatzidis1, 2, E. Harel1† 1

Department of Chemistry, Northwestern University, Evanston, IL 60208, USA

2

Argonne-Northwestern Solar Energy Center, Northwestern University, Evanston, IL 60208, USA †Corresponding author. Email: [email protected] (EH)

Abstract: Metal halide perovskites are promising solar energy materials, but their mechanism of action remains poorly understood. It has been conjectured that energetically stabilized states such as those corresponding to polarons, quasiparticles in which the carriers are dressed with phonons, are responsible for their remarkable photo-physical properties. Yet, no direct evidence of polarons or other low-energy states have been reported despite extensive efforts. Such states should manifest as below bandgap features in transient absorption and photoluminescence measurements. Here, we use single-particle transient absorption microscopy on MAPbI3 (MA= methylammonium) to unambiguously identify spectrally narrow sub-bandgap states directly; we demonstrate that such signals are completely averaged away in ensemble measurements. Carrier temperature dependent studies suggest that hot carriers are directed towards transient low-energy states which are immune from permanent defects and traps, thereby giving rise to low carrier recombination rates, and, ultimately, high power conversion efficiency in devices. The utilization of short-lived sub-bandgap states may be a key design principle that propels widespread use of highly heterogeneous materials in optoelectronic applications. Keyword: Metal halide perovskite, transient absorption microscopy, ultrafast spectroscopy, polaron states, spatially-resolved measurements Halide perovskites are promising materials for a wide range of optoelectronic applications such as photovoltaics and light-emitting diodes1. Their large optical absorption cross sections2, 3, low exciton binding energy4, 5, and high charge carrier mobility6-8 have propelled devices based on perovskites such as methylammonium lead iodide, MAPbI3, to record-setting power conversion efficiencies approaching the performance of silicon based solar cells9, 10. However, the defining

ACS Paragon Plus Environment

1

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 10

characteristics of these materials that give such favorable properties over other inorganic semiconductors remains a contentious topic despite extensive efforts to study the nature of the carriers and their dynamics. It is becoming increasingly clear that heterogeneity in the local morphology, structure, and long-range ordering of these solution-processed materials necessitate an approach that does not average over large spatial domains. Recent work in our lab11 has shown that many of the basic properties of carriers such as the exciton binding energy may vary dramatically on the nanometer length scale. This highlights the importance of studying these materials at a resolution that overcomes spatial disorder, thereby revealing features that would otherwise be averaged away in ensemble measurements. In this work, we use nanoscale ultrafast spectroscopy to directly identify transient sub-bandgap states that have the characteristics of carrier-phonon quasi-particles, or polarons. These transient defects may be thought of as protective states that are immune from permanent defects and traps commonly found in solutionprocessed materials. In order to study thin films of MAPbI3 perovskite with nanometer resolution, we employed a home-built spectrally-resolved transient absorption (TA) microscope. Details of the setup and sample characterization are provided in the SI. Briefly, we use above-bandgap excitation tuned to 2.4 eV and a broadband continuum probe that spans 1.5 – 2.0 eV. The bandgap of MAPbI3 lies around 1.63 eV. Twenty-four individual perovskite locations were selected at random and spectrally-resolved TA spectra were measured at delay times ranging from -0.1 ps to 2 ps. The photoluminescence (PL) spectrum was simultaneously recorded and subtracted from each TA measurement. SEM images (see SI) reveals that the crystalline domains in our thin film sample show two prominent crystal morphologies: small crystallites in the 100 – 300 nm range, and larger, elongated crystallites in the 700 – 1000 nm range. As the spatial resolution of our optical microscope is about 500 nm, most regions selected are from single crystalline domains, while a few have contributions from a small (< 4) number of individual crystallites. In figure 1A and 1B, TA surface plots show the differential transmission, ∆T/T, as a function of the pump-probe delay (vertical axis) and spectrally resolved differential probe transmission (horizontal axis), averaged over all 24 measurements. This sub-ensemble spectrum agrees well with previously published ensemble measurements5, 12, 13 that typically averages over tens to hundreds of microns (thousands of individual crystallites). In general, the TA spectrum shows three prominent features: 1) a short-lived and negative low-energy photo-induced absorption (PIA) feature which is due to bandgap renormalization, 2) a positive bleach feature near the band edge, and 3) a broad, high-energy PIA feature which is due to a combination of continuum absorption and band filling. No excitonic features are apparent in the ensemble-averaged linear absorption spectrum, but as others have demonstrated that even at room temperature, where the binding energy is small, excitonic effects may contribute to the TA spectra12. In Figure 1C and 1D, we display a representative TA spectrum at one specific location (see SI for data from other positions), which is nearly identical to the averaged spectrum. Notably, a new feature (marked with an arrow) is present below the bandgap, near 1.57 eV, that is largely absent in our sub-ensemble averaged spectrum and completely absent in the full-ensemble measurements reported to date. The feature is more clearly visible by examining cuts of the TA surface in time; manifested as a small bump on the low-energy side of the bleach. To isolate this feature, we fit the full TA spectrum at each delay to a band-filling model (see SI), which reproduces nearly all the features except for the sub-bandgap signal. The residual shows a single, isolated peak at 1.572(7) eV with a FWHM of about 36.7(6) meV, which is considerably narrower than other features observed. The red-shift of

ACS Paragon Plus Environment

2

Page 3 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

this feature is approximately 60 meV from the band edge, a value in line with recent theoretical calculations that predict a ~40 meV Frohlich coupling constant between charge carriers and longitudinal optical phonons in lead iodide perovskites at room temperature14. Not all particles showed a prominent sub-bandgap feature, although about 40% did at slightly varying spectral positions and intensities (see SI for additional analysis). The variance in spectral positions is the reason why the ensemble-averaged measurements have not previously observed this state. Next, we analyze the time evolution of the sub-bandgap state and its dependence on initial carrier temperature. To extract the temperature, we measured the TA traces at thirteen different pump intensities. While knowledge of the pump fluence, reflection, and transmission coefficients is, in principle, sufficient to extract the carrier density, we find that local heterogeneity in the MAPbI3 thin films is so extensive that such values are not reliable. Instead of recovering the average carrier densities (which, we estimate to be in the range of 1017 – 1018 cm-3; see SI), we extracted the carrier temperature, which may be determined without relying on the above-mentioned assumptions, by fitting the high-energy tail of the TA spectrum. PL studies show that the carrier density is below the saturation limit, indicating densities far below that needed for complete band-filling (see SI, Figure S7). At each pump power and for each pump-probe delay, we extract a carrier temperature by fitting the TA trace to a band filling model which depends explicitly on the Fermi-Dirac distribution of electrons and holes (see SI for details). This allows for the dependence of the spectral position, intensity, and dynamics of the sub-bandgap state on the initial carrier temperature to be determined. Figure 2A shows temperature maps as a function of average pump fluence and pump-probe delay at two different locations. As expected, the carriers cool rapidly after excitation, and the average carrier temperature increases with pump fluence as the carrier density increases. Note, however, that the initial carrier temperature and cooling rates are not uniform, a consequence of the large degree of heterogeneity in the morphology of thin films of MAPbI3. In Figure 2C and D, we show the isolated sub-bandgap state at two different spatial locations as a function of the initial carrier temperature (measured at T = 240 fs, chosen to establish a quasi-equilibrium by means of carrier-carrier scattering and to avoid unwanted pulse overlap effects). At all positions that exhibit a clear sub-bandgap feature, we observe a small blue shift of the band by an average of 6.1(4) µeV/K, which corresponds to 7.1 meV across the initial carrier temperature range (some positions showed a shift of as much as 18 meV, see SI for analysis), suggesting that this state has excitonic character. As the carrier density increases, carriers may screen the electron-hole interaction causing a blue-shift of the absorption15. In addition, we observe a near-linear relationship between the time-integrated intensity of the peak and inverse temperature (see inset of Figure 2C, D). Shown in Figure 3 are residual TA surfaces for measurements at two separate positions (pos10 and pos4), and at three different initial carrier temperatures. Cuts through the maximum of the sub-bandgap feature show the signal grows rapidly with time. At low initial carrier temperatures, the sub-bandgap feature is weak and its onset appears delayed. As the carrier temperature increases, the feature signal strength increases, while its onset appears earlier in time. For instance, at position 10, the sub-bandgap feature does not appear until after 0.5 ps for an initial carrier temperature of 691 K, while at a carrier temperature of 1196 K, the signal appears within the response time of the instrument (< 200 fs). Complicating the analysis is the imperfect background subtraction by the band filling model and the fact that the carriers cool with time, thereby dynamically changing the rate of formation. This makes fitting to a single rate

ACS Paragon Plus Environment

3

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 10

constant and extracting an activation energy problematic. Nevertheless, an average rise time of about ~180 fs may be extracted, corresponding to half the period of a ~95 cm-1 mode, in close agreement with the frequency of the Pb-I stretch of the inorganic cage in MAPbI3 as measured by Raman spectroscopy16. A deeper theoretical analysis will be needed to identify the exact mechanism by which this new state forms upon carrier cooling, but the observed time scale is generally consistent with lattice distortions that at a minimum require half a vibrational period of the coupled phonon mode to form. In addition to TA, we also carried out fluence-dependent steady-state PL measurements. The ensemble-averaged PL shows a peak near 1.6 eV at room temperature with a FWHM of about 100 meV at room temperature17. In the individual PL measurements, we observe considerable structure not seen in the ensemble spectrum (Figure 4). To better understand these features, we analyzed the PL as a function of the initial carrier temperature. Broadening of the features is most evident away from the PL emission maximum. Since the high-energy tail of the PL emission is described by a Maxwell-Boltzmann distribution, we analyzed the spectra on a logarithmic scale as shown in the inset of the figure. Three distinct regions are observed, labelled I – III in the plot. Regions I and III exhibit clear linear dependence on energy with a large T0 dependence (i.e. variance in the slope), while region II shows the least fluencedependent broadening. The mean and variance of the slopes in all the regions differ substantially, implying that each region represents a unique emissive state. The main emission band in region II is dominant, but the other emissive states clearly contribute. Region III corresponds to hot PL emission. Region II shows negligible carrier temperature dependence because these photons originates from carriers that recombine at the band edge. Notably, region I is hotter than the band-edge emission. However, fitting the low-energy tail to a precise carrier temperature is challenging due to the probe cutoff on the red-side of the band edge. We assign this feature to originating from emission of the sub-bandgap state identified in our TA measurements, exhibiting a redshift of about 50 – 60 meV. Future investigations will focus on measuring the lifetime of each of these emissive states, which has implications for their role in carrier diffusion. In conclusion, we have used nanoscale ultrafast spectroscopy to identify a transient sub-bandgap state in MAPbI3. This state shows many of the hallmarks of a polaronic state, lying about 60 meV below the band edge. The narrow absorption and emission line width suggests that this state has a significantly longer lifetime than states at the center of the Brillouin zone, while timeresolved PL still needs to be performed for additional confirmation. Transient absorption measurements reveal that the rise time of the sub-bandgap signal is on the order of the period of relevant lattice vibrations, further supporting the polaronic picture. A slight blue shift of the subbandgap feature with carrier density suggests that this state is at least partially excitonic in nature. Ensemble averaging obscures these critical features, highlighting the importance of making measurements that overcome spatial and spectral disorder. We believe these results pave the way for future experimental and theoretical work to exploit access to low-energy, protected states that may dramatically improve power conversion efficiency. Data availability. All data associated with this study is available from the corresponding author upon reasonable request.

ACS Paragon Plus Environment

4

Page 5 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Supporting Information. The Supporting Information is available free of charge on the ACS Publications website. Film characterization (S1), details of TA microscope (S2), complete TA datasets (S3), residual sub-bandgap feature and carrier temperature dependence (S4), subbandgap blue-shift statistics (S5), spectrally-resolved PL (S6), and integrated PL dependence on pump power (S7).

Acknowledgments. E.H. acknowledges support by the Air Force Office of Scientific Research (FA9550-14-1-0005), and the Packard Foundation (2013-39272) in part. M.G.K. acknowledges support by the U.S. Department of Energy, Office of Science, Basic Energy Sciences (DE-SC0012541). The electron microscopy work made use of the EPIC facility of Northwestern University’s NUANCE Center, which has received support from the Soft and Hybrid Nanotechnology Experimental (SHyNE) Resource (NSF ECCS-1542205); the MRSEC program (NSF DMR-1121262) at the Materials Research Center; the International Institute for Nanotechnology (IIN); the Keck Foundation; and the State of Illinois, through the IIN. Author contributions. E.H., S.N., and B.S. conceived the ideas and designed the experiments. S.N., B.S., X.J. carried out TAM experiments, and E.H., S.N., and X.J. analyzed the data. C.M.S. and C.S. synthesized the perovskite films and performed the XRD and diffuse reflectance measurements. S.N., X.J., and E.H. contributed to writing the paper. All authors discussed the results in the paper. Competing financial interests. The authors declare no competing financial interests.

ACS Paragon Plus Environment

5

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 10

Figures

Figure 1. a) Transient Absorption (TA) surface averaged over 24 measurements. b) TA spectra at different pump-probe delays. Dashed line is the averaged photoluminescence (PL) spectrum. c) Single-particle TA surface at position 10. Arrow points to subbandgap feature. d) Cuts at 0.36 ps (magenta) and 1.32 ps (green) of single-particle TA spectrum, along with fits (blue curves) and residuals (dashed black curve, scaled).

ACS Paragon Plus Environment

6

Page 7 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2. Temperature maps as a function of pump probe delay and average pump fluence at positions 4 (a) and 20 (b). (c) and (d) Corresponding sub-bandgap feature integrated over all pump-probe delays as a function of the initial carrier temperature. Red points mark center of the peak extracted from a Gaussian fit. Inset: Intensity of peak as a function of inverse initial carrier temperature.

ACS Paragon Plus Environment

7

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 10

Figure 3. Time-dependence of sub-bandgap feature at three different pump fluence values at position 10 (a) and position 4 (b). Black dashed line is the maximum of the signal, while the blue curve is the cut at that spectral location.

ACS Paragon Plus Environment

8

Page 9 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4. Normalized photoluminescence as a function of initial carrier temperature at positions 5 (a) and 11 (b). Inset: logarithm plot of the PL spectra. Black-dashed lines are linear fits to the highest temperature (i.e. highest pump fluence) PL spectrum (red curve) in regions I and III. Filled curves are Gaussian fits to the lowest-energy PL spectrum using constraints set by the regions labelled in the inset.

ACS Paragon Plus Environment

9

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 10

References: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

Gao, P.; Gratzel, M.; Nazeeruddin, M. K. Energ Environ Sci 2014, 7, 2448-2463. Park, N. G. Mater. Today 2015, 18, 65-72. Tan, Z. K.; Moghaddam, R. S.; Lai, M. L.; Docampo, P.; Higler, R.; Deschler, F.; Price, M.; Sadhanala, A.; Pazos, L. M.; Credgington, D.; Hanusch, F.; Bein, T.; Snaith, H. J.; Friend, R. H. Nat Nanotechnol 2014, 9, 687-692. Miyata, A.; Mitioglu, A.; Plochocka, P.; Portugall, O.; Wang, J. T. W.; Stranks, S. D.; Snaith, H. J.; Nicholas, R. J. NatPh 2015, 11, 582-588. D'Innocenzo, V.; Grancini, G.; Alcocer, M. J. P.; Kandada, A. R. S.; Stranks, S. D.; Lee, M. M.; Lanzani, G.; Snaith, H. J.; Petrozza, A. Nat Commun 2014, 5. Motta, C.; El-Mellouhi, F.; Sanvito, S. Sci. Rep. 2015, 5. Johnston, M. B.; Herz, L. M. Acc. Chem. Res. 2016, 49, 146-154. Guo, Z.; Manser, J. S.; Wan, Y.; Kamat, P. V.; Huang, L. B. Nat Commun 2015, 6. Yang, B.; Dyck, O.; Poplawsky, J.; Keum, J.; Puretzky, A.; Das, S.; Ivanov, I.; Rouleau, C.; Duscher, G.; Geohegan, D.; Xiao, K. J. Am. Chem. Soc. 2015, 137, 9210-9213. Son, D.-Y.; Lee, J.-W.; Choi, Y. J.; Jang, I.-H.; Lee, S.; Yoo, P. J.; Shin, H.; Ahn, N.; Choi, M.; Kim, D.; Park, N.-G. Nature Energy 2016, 1, 16081. Nah, S.; Spokoyny, B.; Stoumpos, C.; Soe, C. M. M.; Kanatzidis, M.; Harel, E. Nat Photonics 2017, 11, 285-289. Yang, Y.; Ostrowski, D. P.; France, R. M.; Zhu, K.; van de Lagemaat, J.; Luther, J. M.; Beard, M. C. Nat Photonics 2016, 10, 53-59. Price, M. B.; Butkus, J.; Jellicoe, T. C.; Sadhanala, A.; Briane, A.; Halpert, J. E.; Broch, K.; Hodgkiss, J. M.; Friend, R. H.; Deschler, F. Nat Commun 2015, 6. Wright, A. D.; Verdi, C.; Milot, R. L.; Eperon, G. E.; Perez-Osorio, M. A.; Snaith, H. J.; Giustino, F.; Johnston, M. B.; Herz, L. M. Nat Commun 2016, 7. Peyghambarian, N.; Gibbs, H. M.; Jewell, J. L.; Antonetti, A.; Migus, A.; Hulin, D.; Mysyrowicz, A. Phys. Rev. Lett. 1984, 53, 2433-2436. Ledinsky, M.; Loper, P.; Niesen, B.; Holovsky, J.; Moon, S. J.; Yum, J. H.; De Wolf, S.; Fejfar, A.; Ballif, C. J. Phys. Chem. Lett. 2015, 6, 401-406. Saba, M.; Cadelano, M.; Marongiu, D.; Chen, F. P.; Sarritzu, V.; Sestu, N.; Figus, C.; Aresti, M.; Piras, R.; Lehmann, A. G.; Cannas, C.; Musinu, A.; Quochi, F.; Mura, A.; Bongiovanni, G. Nat Commun 2014, 5.

ACS Paragon Plus Environment

10