Spatially Non-uniform Trap State Densities in ... - ACS Publications

Benjamin J. FoleyShelby CuthriellSina YazdiAlexander Z. ChenStephanie M. .... Lydia H. Manger , Matthew B. Rowley , Yongping Fu , Alexander K. Foote ...
0 downloads 0 Views 2MB Size
Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Letter

Spatially Non-Uniform Trap State Densities in Solution-Processed Hybrid Perovskite Thin Films Sergiu I Draguta, Siddharatha Thakur, Yurii Morozov, Yuanxing Wang, Joseph S. Manser, Prashant V. Kamat, and Masaru Kuno J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.5b02888 • Publication Date (Web): 03 Feb 2016 Downloaded from http://pubs.acs.org on February 4, 2016

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.

The Journal of Physical Chemistry 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 30

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

The Journal of Physical Chemistry Letters

Spatially Non-Uniform Trap State Densities in Solution-Processed Hybrid Perovskite Thin Films Sergiu Draguta1, Siddharatha Thakur1,2, Yurii Morozov1, Yuanxing Wang1, Joseph S. Manser3,4, Prashant V. Kamat 1,3,4, Masaru Kuno1* 1

Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, Indiana 46556, USA 2 Nanotechnology Engineering, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada 3 Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, Indiana 46556, USA 4 Radiation Laboratory, University of Notre Dame, Notre Dame, Indiana 46556, USA *[email protected] Abstract: The facile solution-processability of methylammonium lead halide (CH3NH3PbI3) perovskites has catalyzed the development of inexpensive, hybrid perovskite-based optoelectronics. It is apparent, though, that solution-processed CH3NH3PbI3 films possess local emission heterogeneities, stemming from electronic disorder in the material.

Herein we

investigate the spatially-resolved emission properties of CH3NH3PbI3 thin films through detailed emission intensity versus excitation intensity measurements. These studies enable us to establish the existence of non-uniform trap density variations wherein regions of CH3NH3PbI3 films exhibit effective free carrier recombination while others exhibit emission dynamics strongly influenced by the presence of trap states. Such trap density variations lead to spatially-varying emission quantum yields and correspondingly impact the performance of both methylammonium lead halide perovskite solar cells and other hybrid perovskite-based devices. Of additional note is that the observed spatial extent of the optical disorder extends over length scales greater than that of underlying crystalline domains, suggesting the existence of other factors, beyond grain boundary-related nonradiative recombination channels, which lead to significant intrafilm optical heterogeneities. 1 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 30

Table of Contents (TOC) Graphic:

Key Words: hybrid perovskite, CH3NH3PbI3, heterogeneity, trap states, emission quantum yield. Introduction Methylammonium lead halide perovskites such as MAPbI3 (MA = CH3NH3+) have recently attracted great interest from various facets of optoelectronics research.1,2,3 This stems from their extraordinary photophysical properties, which include direct, near optimal bandgaps and large absorption coefficients –properties which make them uniquely suited for applications such as photovoltaics. Early reports of MAPbI3 solar cells utilized traditional liquid-junction4,5 and solid-state6 sensitized DSSC architectures. By optimizing the composition of hybrid perovskites, corresponding solar cell efficiencies have achieved a certified 21% power conversion energy within a relatively short period of time.7,8,9 The accessible solution-phase synthesis of hybrid perovskites simultaneously enables compositional modification of their absorption and emission properties through halide anion mixing (e.g., MAPbI3-xBrx).10,11

This enables emission

wavelength tuning12 from 760 nm to 530 nm and makes these materials suitable for light 2 ACS Paragon Plus Environment

Page 3 of 30

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

The Journal of Physical Chemistry Letters

emitting13 as well as lasing applications.14,15,16 In addition, ambipolar charge transport along with relatively long carrier diffusion lengths,17,18,19 make lead halide perovskites suitable for other optoelectronic applications such as light emitting field effect transistors.19 A salient advantage of using MAPbI3 in photovoltaics is the significant cost savings achieved through solution processing.20 Today, the most common approach for synthesizing MAPbI3 thin films involves spin coating equimolar mixtures of methylammonium iodide and lead iodide precursors onto substrates followed by post deposition annealing.21 Resulting films, however, exhibit heterogeneous optical and electrical responses due to local compositional, trap state density, and morphological variations. In this regard, compositional heterogeneities may involve nonstoichiometric (e.g. MA or PbI2 rich/poor) regions of the film whereas non-uniform trap state density distributions potentially arise from local differences in film disorder, possibly linked to different average domain sizes and associated number of grain boundaries. Attesting to this, several groups have recently reported local emission heterogeneities in hybrid perovskite thin films.22,23,24,25 Specifically, deQuilettes et al. have observed that MAPbI3 films exhibit shorter lifetimes, lower photoluminescence (PL) intensities, and redshifted emission at grain boundaries relative to the emission from crystalline domains.23 Speculation for why this occurs centers on the possible existence of larger trap state densities at grain boundaries. Simson and co-workers have likewise reported spatially-resolved transient absorption kinetics, showing variable two- versus three-carrier dynamics attributed to localized exciton versus free carrier populations in MAPbI3 films.26

Guo et al. have analogously reported spatially-resolved

variations in carrier diffusion coefficients through local pump/offset probe transient differential absorption measurements.24 These examples all point to the existence of spatial disorder in solution-processed hybrid perovskite films and to the need for more detailed measurements of

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

their optical heterogeneities given the impact that such local disorder has on the overall performance of MAPbI3-based optoelectronics.

As an illustration, local carrier trapping

adversely impacts the performance of MAPbI3 solar cells by reducing both open circuit voltages and fill factors.27,28,29 Herein we investigate the localized optical response of MAPbI3 films using spatially-resolved and correlated emission/absorption measurements.

Variations in emission recombination

dynamics are deduced from measurements of local emission intensity (Iem) versus excitation intensity (Iexc) dependencies wherein such power-dependent measurements provide insight into local carrier kinetics, the existence of carrier trap states and their corresponding densities.30 What results is evidence for large spatial heterogeneities in carrier recombination dynamics under one Sun excitation intensity conditions wherein some regions of MAPbI3 films show predominantly bimolecular radiative recombination while others exhibit optical responses strongly influenced by carrier trapping.

Of importance is that subsequent analyses enable

quantitative estimates of nonuniform trap state distributions in MAPbI3 films as well as corresponding variations in emission quantum yields. We additionally observe that these spatial heterogeneities occur over length scales larger than that associated with the size of underlying crystalline domains. This strongly implies the existence of other factors, beyond the presence of grain boundary-related nonradiative recombination channels, which lead to significant intrafilm optical heterogeneities in hybrid perovskite films. Iexc-dependent emission measurements are useful in revealing relevant recombination pathways for photogenerated carriers in materials30 because emission intensities experimentally  grow in a power law fashion, (i.e.  ∝  ) with power law coefficients (b) that reflect the

underlying carrier recombination processes present.

In the case of exclusive free carrier

4 ACS Paragon Plus Environment

Page 4 of 30

Page 5 of 30

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

The Journal of Physical Chemistry Letters

recombination under continuous wave (CW) excitation, Iem grows linearly with Iexc (i.e.  ∝  , b=1.00). Although excitonic emission also scales linearly with Iexc, this contribution is excluded in MAPbI3 thin films in the absence of unambiguous evidence for the existence of excitons at room temperature.31 In the event that carrier trapping competes with bimolecular electron/hole recombination, the emission growth order changes. Namely, Iem grows as  ∝  / with a power law coefficient of b=1.50.30 Trapping of both photogenerated electrons and holes analogously leads to superlinear (quadratic) growth of Iem (i.e.  ~ ), with a power law

coefficient of b=2.00.30 Observed power law coefficients in Iem versus Iexc plots can therefore be used to infer the operating carrier recombination mechanism in MAPbI3 films, the participation of sub gap states, and, in certain cases, the nature of the excited species (i.e. whether free carriers or excitons). Detailed derivations of these Iexc power law dependencies can be found in the Supporting Information (SI). Figure 1a shows results from typical Iexc-dependent Iem measurements conducted on ~100 nm thick MAPbI3 thin films. Detailed information regarding the thin film preparation can be found in the Experimental Section. The data clearly shows that at relatively low excitation intensities Iem grows superlinearly with excitation intensity. A power law coefficient of b=1.50 is seen over a range of Iexc values at or below 1 Sun (100 mW/cm2, denoted by the shaded region in Figure 1a).

At higher intensities, a linear regime appears and will be explained shortly.

This

superlinear Iem growth in MAPbI3 films has previously32 been attributed to the trapping of photogenerated electrons in MAPbI3 films. Although the exact origin of these trap states is unknown, they are thought to be surface related.25,33 Next, Figure 1a shows that at large excitation intensities (Iexc > 50 W/cm2) the growth order changes with Iem growing linearly with Iexc.

Onset of this behavior is rationalized by the

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

saturation of long-lived electron traps in MAPbI3 films, given reported lifetimes exceeding 100 µs and enhanced bimolecular radiative recombination rates at high Iexc values.34 Recombination kinetics of MAPbI3 films at high excitation intensities are therefore dictated by bimolecular electron-hole recombination in the absence of higher order Auger processes. Of particular relevance is that corresponding trap state densities can be estimated from experimental Iem versus Iexc curves through appropriate kinetic modeling. Thus, by fitting the experimental data to numerical results we find a nominal trap density of Nt=6.32 x 1017 cm-3 in the ~100 nm hybrid perovskite thin film. Figure 1b illustrates the model along with relevant rates for the various carrier recombination processes present. The obtained Nt value is in good agreement with prior trap densities reported in the literature, as determined through power- and time-dependent photoluminescence measurements.23,32,34

Additional information about the

kinetic modeling/fitting involved in extracting trap state densities can be found in the SI. An important consequence of carrier trapping is that hybrid perovskite emission quantum yields (QYs) are variable. To illustrate, the inset of Figure 1a shows that the normalized  emission intensity of MAPbI3 films [i.e.  ⁄



 , where



 =  is a normalized excitation 

 intensity with  a maximum excitation intensity at which a peak QY (  ) is observed] 

grows with increasing excitation intensity due to progressive trap saturation. This implies that QYs increase with Iexc. Upon trap saturation, the external QY plateaus at an optimal value,   ,  and remains constant in the absence of additional non-radiative Auger recombination pathways. Beyond Iexc = 30 W/cm2, Figure 1a shows that  ⁄



 decreases with Iexc and reveals that nonradiative Auger processes increasingly dictate carrier recombination within MAPbI3 films.32,35  The data thus shows that  = 30 W/cm2 is an optimal intensity where (a) carrier trapping is

6 ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30

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

The Journal of Physical Chemistry Letters

minimal due to trap saturation, (b) emission QYs are largest, and (c) carrier densities are low enough such that Auger effects do not significantly influence carrier recombination. Of specific note is that these Iem versus



 measurements can be used to estimate absolute external emission QYs.36 Namely, by modeling the photogenerated carrier recombination in hybrid perovskites in terms of non-radiative carrier trapping, radiative bimolecular recombination and Auger processes it can be shown that  





=  1

!1

  "

#$%



 & '

(

(1)

where  is constant of proportionality. A detailed derivation of Equation 1 can be found in the

Figure 1 (a) Power-dependent emission intensity for a MAPbI3 thin film and a single crystal. Dashed lines are linear fits to the data at low and high excitation intensities. (Inset) Corresponding emission QYs where dashed lines are fits to Equation. (b) Illustration of the trapping model used, which encompasses the following processes: (1) Generation of free carriers (G~αIexc), (2) radiative bimolecular recombination with a second order rate constant kb, (3) trapping of free electrons by trap states with a second order rate constant kt and with a total trap density of Nt (4) Recombination of trapped electrons with available free holes with a second order rate constant kh.

SI.  data to Equation 1, we find a peak emission Consequently, by fitting experimental  ⁄



 QY of   ~70% for the thin film at  = 30 W/cm2. This value is in excellent agreement 

with prior QY values found for MAPbI3 thin films (QY~70%).37 The same analysis finds that hybrid perovskite QYs are on the order of 5.8% at 1 Sun due to carrier trapping. This value is 7 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

again consistent with the prior literature, which reports an estimated 1 Sun external QY of 6.4%.23 For comparison purposes, analogous Iem versus Iexc measurements have been conducted on a bulk MAPbI3 single crystal.

Details about the single crystal growth can be found in the

Experimental Section. Results from these measurements (Figure 1a) show that the overall behavior of the bulk sample is near identical to that of the thin film. The only significant difference is that both the transition of the power law coefficient from b=1.50 to b=1.00 as well as the plateauing of the emission QY occurs at a lower Iexc value. This suggests that the bulk specimen possesses a smaller overall trap density. In fact, modeling the superlinear growth of Iem with Iexc, as done earlier, results in an extracted bulk trap state density of Nt=3.5x1016 cm-3 along with a maximum QY of   ~80% at a correspondingly smaller (optimal) excitation   intensity of  = 7 W/cm2. At 1 Sun, the estimated external QY is ~23%. As before, these

bulk trap state densities and QYs are consistent with what has previously been reported.32,37 Furthermore, the conclusion that bulk specimens possess smaller trap densities than corresponding thin films is supported by reported carrier diffusion lengths. Namely, thin film carrier diffusion lengths are seen to vary from hundreds of nanometers to several microns24,38,39,40 whereas single crystal diffusion lengths readily exceed hundreds of micrometers.41,42 The above Iem versus Iexc measurements therefore lead to quantitative measures of MAPbI3 thin film trap state densities.

When conducted in a spatially-resolved manner, it becomes

possible to extract local trap densities. We have therefore conducted spatially-resolved Iem versus Iexc measurements to obtain two-dimensional (2D) trap density maps of MAPbI3 thin films. At each point (one pixel represents an approximate 600 x 600 nm area), the obtained data is fit to a power law function to extract the corresponding power law coefficient. Given that

8 ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30

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

The Journal of Physical Chemistry Letters

excitation intensities in the range Iexc = 10-2 to 100 W/cm2 have previously been shown to be most relevant in revealing trap state participation in MAPbI3 photogenerated carrier dynamics (Figure 1a), maps have been generated using data acquired over these intensities. Figure 2a represent a typical b coefficient map that results. From the image, it is apparent that sizable b coefficient spatial heterogeneities exist with distinctive variations, ranging from b=1.08 to b=1.52. These variations persist over sizable length scales in the film, on the order of ~5-10 µm. Given that the average crystallite size in these films is 100-200 nm, as determined using scanning electron microscopy (SEM) (Figure S1, SI), the data shows that spatially heterogeneous optical responses exist in MAPbI3 films on length scales larger than underlying domain sizes.

Such long range optical heterogeneities on length scales beyond those of

crystalline domains have not been extensively reported in the literature. Apart from the current investigation, we know of only one other study which shows long range emission intensity heterogeneities in MAPbI3 and MAPbI3-xBrx thin films.43 It is evident from the above power law analysis that regions of the film having b values close to b=1.50 represent areas where significant carrier trapping occurs and where trap densities are large (i.e. >1017 cm-3). Conversely, regions where b values are closer to b=1.00 indicate areas of predominant bimolecular electron-hole recombination and where trap densities are correspondingly smaller (i.e. ~1016 cm-3). Figure 2b highlights this, showing local Iem versus Iexc plots for two specific regions of the 2D image in Figure 2a (denoted by the circled points). Associated with each curve is a corresponding trap density, estimated via the fitting procedure outlined earlier. To ensure that trap saturation, first seen in Figure 1, can be achieved, analogous b coefficient maps have been generated using excitation intensities in the range Iexc = 101-103

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 30

W/cm2. Resulting maps (Figure S2, SI) clearly reveal nearly homogenous b=1.00 images in contrast to what is seen at lower intensities. Through additional analyses, b-coefficient images have been converted to corresponding Nt maps. Figure 2c shows that resulting trap densities range from 9.0 x 1015 to 8.6 x 1017 cm-3 with an average density of ~1.1 x 1017 cm-3. Clear order of magnitude Nt differences exist and corroborate the substantial electronic disorder previously seen in hybrid perovskite films.25 Details regarding the conversion of b coefficient maps to Nt maps can be found in the SI. This trap density variability leads to commensurately variable local emission QYs. Figure 2d shows a QY map of the same area imaged in Figure 2a. When analyzed, it is apparent that low QY regions directly correspond to trap rich sections of the film (i.e. b~1.50). Conversely, high QY regions directly map onto sections of the film that predominantly exhibit bimolecular electron-hole recombination (i.e. b~1.00). Details regarding how QY maps were generated can

Figure 2. (a) b coefficient map extracted from spatially-resolved Iem versus Iexc measurements. (b) Iem 10 versus Iexc traces from two different sections on the film as indicated. (Inset) TCSPC decay from high and low PL regions of the film as indicated. TrapEnvironment density map extracted from fitting Iem versus Iexc ACS Paragon(c) Plus measurements. (d) Quantum yield map extracted at Iexc=0.1W/cm2 for the same area. The length of the scale bar is equal to 10 µm.

Page 11 of 30

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

The Journal of Physical Chemistry Letters

be found in the SI. This direct correlation is further established using an image pixel correlation plot shown in Figure S3 of the SI. We conclude that, as expected, trap rich (poor) regions exhibit lower (higher) emission QYs at one Sun due to competitive (non-competitive) carrier trapping. The above conclusion is also reflected in carrier lifetimes observed from trap rich and trap poor regions. Namely, local emission lifetimes have been obtained through time-correlated single photon counting (TCSPC) measurements of photoluminescence decays acquired from different sections of films. The data in the inset of Figure 2b shows TCSPC decays taken from the circled areas in Figures 2a, 2c, and 2d. It is evident that decays from trap rich regions are shorter than those from trap poor regions. Fits show average 1/e lifetimes of τ1/e=16.4±0.1 ns (τ1/e=24.3±0.1 ns) (10 µJ/cm2/pulse) for the trap rich (poor) region, in qualitative agreement with expectations from competitive carrier trapping.

All of the acquired results, the b coefficient

maps, the spatially-resolved QYs, and corresponding lifetimes, are thus self-consistent with the existence of heterogeneous trap densities across MAPbI3 films.

Figure 3. a) Absorption normalized photoluminescence map. b) Corresponding absorptance map acquired at λexc= 520 nm (Iexc = 0.05 W/cm2, CW). The length of the scale bar is equal to 10 µm.

11 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 12 of 30

To correlate local b coefficient/trap state density and QY variations with other spatialheterogeneities in MAPbI3 optical response, we have conducted correlated integrated emission intensity mapping, absorption imaging and spectral imaging of the same regions previously analyzed. Specifically, Figure 3a shows an absorption-normalized, integrated emission map of the same region imaged in Figure 2a (Iexc = 0.05 W/cm2, λexc= 520 nm, CW). As expected from the above QY analysis, substantial spatial heterogeneities exist in emission intensities across the film with Iem variations exceeding 30%.23,24,25,26 These emission intensity images have been normalized by the absorption of the film in order to exclude the effects of any thickness variations. Figure 3b shows the corresponding 520 nm absorptance map used to carry out the normalization. From the image, it is clear that slight thickness variations exist across the film. To quantify them, an absorption coefficient of α520 nm=1.12

x 105 cm-1

44

at 520 nm was used to establish film thicknesses, yielding an average

thickness of 102 nm along with a thickness variation of 4 nm. The latter value agrees with results from separate atomic force microscopy (AFM) measurements we have conducted, which reveals an average surface roughness on the order of 10 nm (Figure S4, SI). Given the ability to compare absorption images to corresponding Nt maps, we have additionally analyzed any correlations between the two and find no link between thicker/thinner areas of the film and regions possessing higher/lower trap densities. A pixel correlation plot demonstrating this can be found in the SI (Figure S5). At this point, to address the existence of possible spectroscopic heterogeneity within MAPbI3 films, we have conducted localized, emission spectrum mapping of same regions imaged to obtain b coefficient and Nt maps. These measurements entail acquiring emission spectra at each point of a 2D map (Iexc = 0.05 W/cm2, λexc = 520 nm, CW) from where emission peak energies at

12 ACS Paragon Plus Environment

Page 13 of 30

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

The Journal of Physical Chemistry Letters

each point are extracted via fitting. Figure 4a illustrates the resulting spectral map, obtained for the same region imaged in Figures 2a, 2c, and 2d. The data show emission peak wavelengths varying over a small range between λem = 762.5 nm (1.626 eV) and λem = 765.2 nm (1.620 eV). The average energy is found to be consistent with the estimated band gap of MAPbI3 for a ~100 nm thick film (Eg ~ 1.63 eV).45 No gross spectral differences are seen in any of the imaged regions.

Figure 4. (a) PL emission peak position map. False color scale unit: nm. (b) Emission spectra from low and high emission intensity regions of the film, as denoted by the circled points. The length of the scale bar is equal to 10 µm.

Figure 4b shows corresponding emission spectra from the two (circled) regions in Figure 2a. These sections of the film have previously been shown to possess dramatically different trap densities as indicated in Figure 2b. Despite the large Nt variation, though, the observed spectra are quite similar. If anything, a slight redshift appears in the spectrum associated with the trap rich region. To assess whether spectral variations in Figure 4a are indeed linked to the trap density variations observed in Figure 2a, a more detailed pixel correlation analysis was conducted. Results from this analysis reveal no correlation and are shown in Figure S6 of the SI. 13 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 14 of 30

Trap rich regions are thus not linked to redshifts in the emission. What is more likely responsible for the observed spectral differences then is reabsorption of emitted light from MAPbI3 films.46 This is corroborated by comparing Figures 4a and 3b where the latter image shows the 520 nm absorptance image of the same region being analyzed. Redshifted regions of the film evidently correspond to thicker sections of the film. This correspondence is made clearer in a subsequent pixel correlation plot shown in Figure S7 of the SI. Additionally, similar redshifts and broadening of PL spectra due to reabsorption effects have been observed by other groups.32,47 We therefore conclude that, at present, no link exists between the intrafilm disorder, which results in the observed trap state density variations, and the spectral response of MAPbI3 thin films. At this point, the above spatially-resolved emission measurements as well as literature23,24,25,26 suggest that optical heterogeneity is a universal aspect of presently made MAPbI3 films. From power-dependent emission measurements, we have established that these heterogeneities stem from large variations in local trap state densities. Consequently, under excitation intensities similar to one Sun illumination, the recombination mechanism experienced by photogenerated species within MAPbI3 films varies from trap-mediated to what is nearly band-to-band recombination. This subsequently impacts the performance of MAPbI3-based optoelectronics. Namely, within the context of MAPbI3 solar cells, power conversion efficiencies should exhibit sizable spatial heterogeneities.

The above analysis additionally suggests that average trap

densities within MAPbI3 films should reside below 9.0 x 1015 cm-3 if improvements to existing power conversion efficiencies are to be realized. All of the above point to the importance of developing new synthetic approaches to minimize traps within hybrid perovskite films as well as to the need for more insight into their actual origin.

14 ACS Paragon Plus Environment

Page 15 of 30

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

The Journal of Physical Chemistry Letters

Conclusions Local emission measurements of hybrid perovskite films have revealed the existence of significant spatial heterogeneities in their optical response. These intrafilm heterogeneities arise from non-uniform trap state density distributions, which extend over significant length scales in the film. Associated trap densities range from 9.0 x 1015 to 8.6 x 1017 cm-3, and result in regions of the film which exhibit near-exclusive bimolecular free carrier recombination while others exhibit trap-mediated recombination at 1 Sun excitation intensities. Corresponding differences also arise in local emission QYs and carrier lifetimes where trap rich (poor) regions exhibit smaller (larger) QYs and shorter (longer) lifetimes. Although the origin of these carrier traps in hybrid perovskites is still unknown, knowledge of their presence and local density is important since this disorder ultimately limits the performance of corresponding hybrid perovskite-based optoelectronics. In this regard, a better understanding and ultimately control over relevant trap states densities will be critical to eventually realizing the full potential of hybrid perovskite solar cells.

Experimental Section MAPbI3 thin film preparation. 148.7 mg of PbI2 (Sigma Aldrich, 99.9%) was dissolved in 700 µL of γ-butyrolactone (Sigma Aldrich, >99%) to create a 0.1 M solution. 52.3 mg of CH3NH3I (Sigma Aldrich, 99.9%) was likewise dissolved in 300 µL of dimethyl sulfoxide (Sigma Aldrich, >99%) to create a 0.1 M solution. The two solutions were then mixed together for 30 minutes at 70 ˚C in a glove box under inert atmosphere. In parallel, glass coverslips were cleaned in ethanol (Sigma Aldrich, 90%) for 30 minutes followed by a 5 minute plasma cleaning step. Perovskite films were made by spin coating the mixed PbI2/CH3NH3I solution onto cleaned

15 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

coverslips at 2000 rpm for 30 seconds.

Page 16 of 30

Drops of toluene were subsequently added while

spinning to solvent anneal the film. Following this step, films were annealed at 110 ˚C for 30 minutes under N2 in a tube furnace. Finally, a thin protective layer of poly(methyl-methacrylate) (PMMA) [Sigma Aldrich, Mw=120,000] was spin coated onto hybrid perovskite films at 2000 rpm for 30 seconds followed by a 10 minute annealing step at 110 ˚C under N2.

X-ray

diffraction (XRD) powder patterns were acquired to verify the formation of the hybrid perovskite. Representative powder patterns can be found in the SI (Figure S8).

Single crystal growth. Single MAPbI3 crystals were prepared according to previous reports.48 The crystallization solution consisted of 922 mg PbI2 and 318 mg CH3NH3I dissolved in 2 mL γbutyrolactone. The solution was heated to 70 °C for 1 h to fully dissolve the precursors and was subsequently filtered using a 0.2 µm PTFE syringe filter to remove any residual particulates. The vial containing the solution was then submerged in an oil bath at 110 °C and was allowed to stand for 3 h. Large single crystals (typically on the order of 4 x 4 x 2 mm) were then removed from solution and were dried under vacuum for 24 h.

Iem versus Iexc measurements and b coefficient mapping. For Iem versus Iexc measurements, emission intensities were collected as a function of excitation intensity using a 520 nm CW laser [Coherent, Obis] focused onto samples using a high numerical aperture objective (0.95 NA, Nikon). The resulting emission was collected with the same objective with a barrier filter (Chroma, HQ 680LP) to reject the laser. The emission was subsequently detected with a single photon counting avalanche photodiode (Perkin Elmer, SPCM AQR-14) using variably crossed, absorptive linear polarizers mounted on rotation stages to adjust the excitation fluence. For b

16 ACS Paragon Plus Environment

Page 17 of 30

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

The Journal of Physical Chemistry Letters

coefficient maps, Iexc was ramped at every point of the image and the emission intensity was monitored with the avalanche photodiode. Individual curves were then fit to a power law function to extract associated power law coefficients. Samples were scanned point by point using an open loop stepper motor stage to create an image. Individual step sizes were 600 nm as determined using a calibration grid (Edmund Optics).

Time-correlated single photon counting. MAPbI3 excited-state lifetimes were taken on the same homebuilt microscope using a pulsed 405 nm diode laser to excite samples. The excitation pulse width was 70 ps at a repetition rate of 2.5 MHz (PicoQuant LDH-P-C 405). Variably crossed, absorptive linear polarizers were used to attenuate the intensity of the source. The resulting emission was passed through a barrier filter (Chroma, HQ 680LP) and was detected with a single photon counting avalanche photodiode (Perkin Elmer, SPCM AQR-14). The output of the detector was subsequently fed into a time-correlated single-photon counter (PicoQuant, PicoHarp 300) to generate decay traces. Lifetime measurements were taken at 10 µJ/cm2/pulse.

Integrated emission intensity mapping. Integrated emission intensity mapping of MAPbI3 films was conducted using the above described homebuilt microscope. The excitation intensity in these measurements was Iexc=0.05 W/cm2.

Spatially-resolved emission spectral measurements. Spatially-resolved emission spectral mapping was conducted using the above described optical microscope. Instead of detecting the emitted light from samples with an avalanche photodiode, the emission was sent to a fiber-based

17 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 18 of 30

spectrometer (Ocean Optics), enabling local emission spectra to be acquired on a point by point basis. Local emission spectra at each point were then fit with a Gaussian function using a home written computer program to extract emission peak positions.

Absorption mapping. Transmission measurements were acquired on the same homebuilt optical setup used for emission measurements. A second objective (NA=0.65, Zeiss), oriented in a collinear geometry to the first objective, was used to collect any transmitted light through samples.

This light was then focused onto a near shot noise limited photodiode (PH200,

Highland Technology). A second PH200 was used as a reference to null out any incident laser power fluctuations during measurements. Referencing was achieved by splitting the incident light prior to the sample with a 50/50 beamsplitter. Half of the excitation was sent through the sample while the other half was sent to the reference photodiode.

To estimate absolute

transmittance values (T), the ratio of transmitted light through samples and a reference intensity, measured through substrates without perovskite films, was used. subsequently estimated as * ≈ 1

Absorptances were

,, assuming a minimal reflectance contribution.

AFM imaging. Atomic force microscopy images of hybrid perovskite films were obtained in non-contact mode using a commercial AFM system (Park Systems, XE-70). Samples were prepared as described previously except that silicon wafers were used as substrates and no protective PMMA layer was deposited over perovskite films.

Powder X-ray Diffraction. A Bruker D8 Advance Davinci powder X-ray diffractometer was used for structural characterization of CH3NH3PbI3 thin films.

18 ACS Paragon Plus Environment

Page 19 of 30

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

The Journal of Physical Chemistry Letters

Acknowledgements: We thank the Office of Basic Energy Sciences of the US Department of Energy through grant DE-SC0014334 and the Center for Sustainable Energy at Notre Dame (ND Energy) for financial support of this study. We also thank Notre Dame Radiation Laboratory for use of its facilities (NDRL 5103). JSM acknowledges the support of King Abdullah University of Science and Technology (KAUST) through award OCRF-2014-CRG3-2268. We thank the ND Energy Materials Characterization Facility (MCF) for the use of the Bruker D8 Advance Davinci powder X-ray diffractometer. The MCF is funded by the Sustainable Energy Initiative (SEI), which is part of the Center for Sustainable Energy at Notre Dame (ND Energy).

Supporting information Detailed derivation of Iexc power law dependencies. Details of the kinetic modeling/fitting used to extract Nt. Derivation of Equation 1. SEM images of MAPbI3 films. High Iexc b coefficient maps. Details of QY map generation. Pixel correlation plot for Nt and QY maps. AFM image of MAPbI3 films. Pixel correlation plot for Nt and A maps. Pixel correlation plot for Nt and emission peak wavelength maps. Pixel correlation plot for emission peak wavelength and A maps. XRD powder patterns of MAPbI3 films. This information can be found on the internet at http:pubs.acs.org

19 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 20 of 30

References (1) Green, M.; Ho-Baillie, A.; Snaith, H. J. The Emergence of Perovskite Solar Cells. Nat. Photonics 2014, 8, 506–514. (2) Stranks, S. D.; Snaith, H. J. Metal-Halide Perovskites for Photovoltaic and Light-Emitting Devices. Nat. Nanotechnol. 2015, 10, 391–402. (3) Sum, T. C.; Mathews, N. Advancements in Perovskite Solar Cells: Photophysics Behind the Photovoltaics. Energy Environ. Sci. 2014, 7, 2518–2534. (4) Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal Halide Perovskites as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 2009, 131, 6050–6051. (5) 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. (6) 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-SolidState Submicron Thin Film Mesoscopic Solar Cell with Efficiency Exceeding 9%. Sci. Rep. 2012, 2, 591. (7) Yang, W. S.; Noh, J. H.; Jeon, N. J.; Kim, Y. C.; Ryu, S.; Seo, J.; Seok, S. I. HighPerformance Photovoltaic Perovskite Layers Fabricated Through Intramolecular Exchange. Science 2015, 348, 1234−1237. (8) Jeon, N. J.; Noh, J. H.; Kim, Y. C.; Yang, W. S.; Ryu, S.; Seok, S. Il. Solvent Engineering for High-Performance Inorganic-Organic Hybrid Perovskite Solar Cells. Nat. Mater. 2014, 13, 1–7.

20 ACS Paragon Plus Environment

Page 21 of 30

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

The Journal of Physical Chemistry Letters

(9) Research Cell Efficiency Records, NREL. Retrieved from: http://www.nrel.gov/ncpv/. Accessed at: 01-25-2016 (10) Fedeli, P.; Gazza, F.; Calestani, D.; Ferro, P.; Besagni, T.; Zappettini, A.; Calestani, G.; Marchi, E.; Ceroni, P.; Mosca, R. Influence of the Synthetic Procedures on the Structural and Optical Properties of Mixed-Halide (Br, I) Perovskite Films. J. Phys. Chem. C 2015, 119, 21304–21313. (11) Noh, J. H.; Im, S. H.; Heo, J. H.; Mandal, T. N.; Seok, S. Il. Chemical Management for Colorful, Efficient, and Stable Inorganic-Organic Hybrid Nanostructured Solar Cells. Nano Lett. 2013, 13, 1764–1769. (12) Tan, Z.-K.; Moghaddam, R. S.; Lai, M. L.; Docampo, P.; Higler, R.; Deschler, F.; Price, M.; Sadhanala, A.; Pazos, L. M.; Credgington, D.; et al. Bright Light-Emitting Diodes Based on Organometal Halide Perovskite. Nat. Nanotechnol. 2014, 9, 687–692. (13) Li, G.; Tan, Z.-K.; Di, D.; Lai, M. L.; Jiang, L.; Lim, J. H.-W.; Friend, R. H.; Greenham, N. C. Efficient Light-Emitting Diodes Based on Nano-Crystalline Perovskite in a Dielectric Polymer Matrix. Nano Lett. 2015, 15, 2640–2644. (14) Stranks, S. D.; Snaith, H. J. Metal-Halide Perovskites for Photovoltaic and Light-Emitting Devices. Nat. Nanotechnol. 2015, 10, 391–402. (15) Fu, A.; Yang, P. Organic-Inorganic Perovskites: Lower Threshold for Nanowire Lasers. Nat. Mater. 2015, 14, 557–558. (16) Xing, G.; Mathews, N.; Lim, S. S.; Yantara, N.; Liu, X.; Sabba, D.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C. Low-Temperature Solution-Processed Wavelength-Tunable Perovskites for Lasing. Nat. Mater. 2014, 13, 476–480.

21 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 22 of 30

(17) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J. Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite Absorber. Science 2013, 342, 341–344. (18) Ponseca, C. S.; Savenije, T. J.; Abdellah, M.; Zheng, K.; Yartsev, A.; Pascher, T.; Harlang, T.; Chabera, P.; Pullerits, T.; Stepanov, A.; et al. Organometal Halide Perovskite Solar Cell Materials Rationalized: Ultrafast Charge Generation, High and Microsecond-Long Balanced Mobilities, and Slow Recombination. J. Am. Chem. Soc. 2014, 136, 5189–5192. (19) Chin, X. Y.; Cortecchia, D.; Yin, J.; Bruno, A.; Soci, C. Lead Iodide Perovskite LightEmitting Field-Effect Transistor. Nat. Commun. 2015, 6, 7383. (20) Snaith, H. J. Perovskites: The Emergence of a New Era for Low-Cost, High-Efficiency Solar Cells. J. Phys. Chem. Lett. 2013, 4, 3623–3630. (21) Tian, Y.; Merdasa, A.; Peter, M.; Abdellah, M.; Zheng, K.; Ponseca, C. S.; Pullerits, T.; Yartsev, A.; Sundström, V.; Scheblykin, I. G. Giant Photoluminescence Blinking of Perovskite Nanocrystals Reveals Single-Trap Control of Luminescence. Nano Lett. 2015, 15, 1603–1608. (22) Nie, W.; Tsai, H.; Asadpour, R.; Neukirch, A. J.; Gupta, G.; Crochet, J. J.; Chhowalla, M.; Tretiak, S.; Alam, M.; Wang, H. High-Efficiency Solution-Processed Perovskite Solar Cells with Millimeter-Scale Grains. Science 2015, 347, 522–525. (23) deQuilettes, D. W.; Vorpahl, S. M.; Stranks, S. D.; Nagaoka, H.; Eperon, G. E.; Ziffer, M. E.; Snaith, H. J.; Ginger, D. S. Impact of Microstructure on Local Carrier Lifetime in Perovskite Solar Cells. Science 2015, 348, 683–686. (24) Guo, Z.; Manser, J. S.; Wan, Y.; Kamat, P. V.; Huang, L. Spatial and Temporal Imaging of Long-Range Charge Transport in Perovskite Thin Films by Ultrafast Microscopy. Nat. Commun. 2015, 6, 7471. 22 ACS Paragon Plus Environment

Page 23 of 30

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

The Journal of Physical Chemistry Letters

(25) Vrućinić, M.; Matthiesen, C.; Sadhanala, A.; Divitini, G.; Cacovich, S.; Dutton, S. E.; Ducati, C.; Atatüre, M.; Snaith, H.; Friend, R. H.; et al. Local Versus Long-Range Diffusion Effects of Photoexcited States on Radiative Recombination in Organic-Inorganic Lead Halide Perovskites. Adv. Sci. 2015, DOI: 10.1002/advs.201500136. (26) Simpson, M. J.; Doughty, B.; Yang, B.; Xiao, K.; Ma, Y.-Z. Spatial Localization of Excitons and Charge Carriers in Hybrid Perovskite Thin Films. J. Phys. Chem. Lett. 2015, 6, 3041–3047. (27) Blakesley, J. C.; Neher, D. Relationship between Energetic Disorder and Open-Circuit Voltage in Bulk Heterojunction Organic Solar Cells. Phys. Rev. B 2011, 84, 12. (28) Yang, W.; Yao, Y.; Wu, C.Q. Origin of the High Open Circuit Voltage in Planar Heterojunction Perovskite Solar Cells: Role of the Reduced Bimolecular Recombination. J. Appl. Phys. 2015, 117, 095502. (29) Chen, B.; Qiao, X.; Liu, C.-M.; Zhao, C.; Chen, H.-C.; Wei, K.-H.; Hu, B. Effects of Bulk and Interfacial Charge Accumulation on Fill Factor in Organic Solar Cells. Appl. Phys. Lett. 2013, 102, 193302. (30) Vietmeyer, F.; Frantsuzov, P. A.; Janko, B.; Kuno, M. Carrier Recombination Dynamics in Individual CdSe Nanowires. Phys. Rev. B 2011, 83, 115319. (31) Miyata, A.; Mitioglu, A.; Plochocka, P.; Portugall, O.; Wang, J. T.-W.; Stranks, S. D.; Snaith, H. J.; Nicholas, R. J. Direct Measurement of the Exciton Binding Energy and Effective Masses for Charge Carriers in Organic-Inorganic Tri-Halide Perovskites. Nat. Phys. 2015, 11, 582–587.

23 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 24 of 30

(32) Saba, M.; Cadelano, M.; Marongiu, D.; Chen, F.; Sarritzu, V.; Sestu, N.; Figus, C.; Aresti, M.; Piras, R.; Geddo Lehmann, A.; et al. Correlated Electron-Hole Plasma in Organometal Perovskites. Nat. Commun. 2014, 5, 5049. (33) Wu, X.; Trinh, M. T.; Niesner, D.; Zhu, H.; Norman, Z.; Owen, J. S.; Yaffe, O.; Kudisch, B. J.; Zhu, X. Trap States in Lead Iodide Perovskites. J. Am. Chem. Soc. 2015, 137, 2089–2096. (34) Stranks, S. D.; Burlakov, V. M.; Leijtens, T.; Ball, J. M.; Goriely, A.; Snaith, H. J. Recombination Kinetics in Organic-Inorganic Perovskites: Excitons, Free Charge, and Subgap States. Phys. Rev. Appl. 2014, 2, 034007. (35) Yang, Y.; Yang, M.; Li, Z.; Crisp, R.; Zhu, K.; Beard, M. C. Comparison of Recombination Dynamics in CH3NH3PbBr3 and CH3NH3PbI3 Perovskite Films: Influence of Exciton Binding Energy. J. Phys. Chem. Lett. 2015, 6, 4688- 4692. (36) Wang, C.; Li, C.-Y.; Hasselbeck, M. P.; Imangholi, B.; Sheik-Bahae, M. Precision, AllOptical Measurement of External Quantum Efficiency in Semiconductors. J. Appl. Phys. 2011, 109, 93108. (37) Deschler, F.; Price, M.; Pathak, S.; Klintberg, L. E.; Jarausch, D.-D.; Higler, R.; Hüttner, S.; Leijtens, T.; Stranks, S. D.; Snaith, H. J.; et al. High Photoluminescence Efficiency and Optically Pumped Lasing in Solution-Processed Mixed Halide Perovskite Semiconductors. J. Phys. Chem. Lett. 2014, 5, 1421–1426. (38) Ponseca, C. S.; Savenije, T. J.; Abdellah, M.; Zheng, K.; Yartsev, A.; Pascher, T.; Harlang, T.; Chabera, P.; Pullerits, T.; Stepanov, A.; et al. Organometal Halide Perovskite Solar Cell Materials Rationalized: Ultrafast Charge Generation, High and Microsecond-Long Balanced Mobilities, and Slow Recombination. J. Am. Chem. Soc. 2014, 136, 5189–5192.

24 ACS Paragon Plus Environment

Page 25 of 30

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

The Journal of Physical Chemistry Letters

(39) Xing, G.; Mathews, N.; Sun, S.; Lim, S. S.; Lam, Y. M.; Grätzel, M.; Mhaisalkar, S.; Sum, T. C. Long-Range Balanced Electron- and Hole-Transport Lengths in Organic-Inorganic CH3NH3PbI3. Science 2013, 342, 344–347. (40) Stranks, S. D.; Eperon, G. E.; Grancini, G.; Menelaou, C.; Alcocer, M. J. P.; Leijtens, T.; Herz, L. M.; Petrozza, A.; Snaith, H. J. Electron-Hole Diffusion Lengths Exceeding 1 Micrometer in an Organometal Trihalide Perovskite Absorber. Science 2013, 342, 341–344. (41) Shi, D.; Adinolfi, V.; Comin, R.; Yuan, M.; Alarousu, E.; Buin, A.; Chen, Y.; Hoogland, S.; Rothenberger, A.; Katsiev, K.; et al. Low Trap-State Density and Long Carrier Diffusion in Organolead Trihalide Perovskite Single Crystals. Science 2015, 347, 519–522. (42) Dong, Q.; Fang, Y.; Shao, Y.; Mulligan, P.; Qiu, J.; Cao, L.; Huang, J. Electron-Hole Diffusion Lengths > 175 µm in Solution-Grown CH3NH3PbI3 Single Crystals. Science 2015, 347, 967–970. (43) Sutter-Fella, C. M.; Li, Y.; Amani, M.; Ager, J. W.; Toma, F. M.; Yablonovitch, E.; Sharp, I. D.; Javey, A. High Photoluminescence Quantum Yield in Band Gap Tunable Bromide Containing Mixed Halide Perovskites. Nano Lett. 2015, 16, 800–806. (44) De Wolf, S.; Holovsky, J.; Moon, S.-J.; Löper, P.; Niesen, B.; Ledinsky, M.; Haug, F.-J.; Yum, J.-H.; Ballif, C. Organometallic Halide Perovskites: Sharp Optical Absorption Edge and Its Relation to Photovoltaic Performance. J. Phys. Chem. Lett. 2014, 5, 1035–1039. (45) D’Innocenzo, V.; Srimath Kandada, A. R.; De Bastiani, M.; Gandini, M.; Petrozza, A. Tuning the Light Emission Properties by Band Gap Engineering in Hybrid Lead Halide Perovskite. J. Am. Chem. Soc. 2014, 136, 17730–17733.

25 ACS Paragon Plus Environment

The Journal of Physical Chemistry 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 26 of 30

(46) Mettler, K. Effect of Reabsorbed Recombination Radiation on Photoluminescence and Photoconductivity in a Semi-Infinite Direct-Gap Semiconductor. Phys. Status Solidi 1978, 49, 163–168. (47)

Tian, Y.; Merdasa, A.; Unger, E.; Abdellah, M.; Zheng, K.; McKibbin, S.; Mikkelsen, A.;

Pullerits, T.; Yartsev, A.; Sundström, V.; et al. Enhanced Organo-Metal Halide Perovskite Photoluminescence from Nanosized Defect-Free Crystallites and Emitting Sites. J. Phys. Chem. Lett. 2015, 6, 4171–4177. (48) Saidaminov,

M. I.; Abdelhady, A. L.; Murali, B.; Alarousu, E.; Burlakov, V. M.; Peng, W.;

Dursun, I.; Wang, L.; He, Y.; Maculan, G.; et al. High-Quality Bulk Hybrid Perovskite Single Crystals Within Minutes by Inverse Temperature Crystallization. Nat. Commun. 2015, 6, 7586.

26 ACS Paragon Plus Environment

Page 27 of 30

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

The Journal of Physical Chemistry Letters

Figure 1 82x34mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

Figure 2 78x62mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

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

The Journal of Physical Chemistry Letters

Figure 3 83x34mm (300 x 300 DPI)

ACS Paragon Plus Environment

The Journal of Physical Chemistry 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

Figure 4 78x31mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 30 of 30