Photoluminescence Flickering and Blinking of Single CsPbBr3

Nov 30, 2018 - To obtain an in-depth understanding of the dynamics and mechanism of carrier recombination in CsPbBr3 nanocrystals (NCs), we have ...
0 downloads 0 Views 951KB Size
Subscriber access provided by University of Winnipeg Library

Surfaces, Interfaces, and Catalysis; Physical Properties of Nanomaterials and Materials 3

Photoluminescence Flickering and Blinking of Single CsPbBr Perovskite Nanocrystals: Revealing Explicit Carrier Recombination Dynamics Sudipta Seth, Tasnim Ahmed, and Anunay Samanta J. Phys. Chem. Lett., Just Accepted Manuscript • Publication Date (Web): 30 Nov 2018 Downloaded from http://pubs.acs.org on November 30, 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 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 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.

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 23 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

Photoluminescence Flickering and Blinking of Single CsPbBr3 Perovskite Nanocrystals: Revealing Explicit Carrier Recombination Dynamics

Sudipta Seth, Tasnim Ahmed and Anunay Samanta School of Chemistry, University of Hyderabad, Hyderabad 500046, India Corresponding Author *Anunay Samanta. E-mail: [email protected]

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 23

ABSTRACT In order to obtain an in-depth understanding of the dynamics and mechanism of carrier recombination in CsPbBr3 nanocrystals (NCs), we have investigated the photoluminescence (PL) of this material at the single particle level using time-tagged-time-resolved method. The study reveals two distinct types of PL fluctuations of the NCs, which are assigned to flickering and blinking. The flickering is found to be due to excess surface trap on the NCs and the flickering single particles are transformed into blinking ones with significant enhancement of PL intensity and stability on post-synthetic surface treatment. Intensity correlated lifetime analysis of the PL time-trace reveals both trap-mediated nonradiative band edge carrier recombination and positive trion recombination in single NCs. Dynamical and statistical analysis suggests a diffusive nature of the trap states to be responsible for the PL intermittency of the system. These findings throw light on the nature of the trap states, reveal the manifestation of these trap states in PL fluctuation and provide an effective way to control the dynamics of CsPbBr3 NCs. TOC GRAPHIC

2 ACS Paragon Plus Environment

Page 3 of 23 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

Semiconductor nanocrystals (NCs) are the most important candidates driving the recent progress in photovoltaic and optoelectronics.1 As photogenerated charge carriers in these NCs are confined to a small volume, strong coulombic interaction between them often drives many interesting multicarrier processes such as multiexciton generation,2,

3

Auger recombination,4,

5

efficient intraband relaxation6 and spectral diffusion.7 Among these, Auger recombination, in which electron - hole recombination energy, instead of being released in the form of a photon, is transferred to a third carrier, is an undesired process for optoelectronic applications.4, 5 The Auger process, which generally occurs in sub-nanosecond time scale in small size NCs, is faster than the radiative lifetime of a single electron-hole (e-h) pair.4 This process assumes significance when multiexcitons are generated (simultaneously) in single NC or when the NCs are charged by direct trapping of one of the charge carriers in a long-lived defect state.8 Quenching of photoluminescence (PL) due to nonradiative Auger recombination in charged NCs introduces intermittency in the single NC PL trajectory.9-11 The simplest form of a charged NC is trion, comprising an exciton and one extra charge.12 Positive (2h-e) and negative (h-2e) trions are responsible for non-emissive or less-emissive off-states of single NC PL intermittency or blinking.12,

13

The NCs must be charged for trion Auger recombination, but charging is not

necessary for blinking.14-16 For example, short-lived shallow trap states can facilitate rapid nonradiative relaxation of the photogenerated carrier before another exciton is generated. This nonradiative process, which does not involve the long-lived traps, is termed in our discussion as nonradiative band edge carrier (NBC) recombination.17 The NBC and trion recombination rates fluctuate with time depending on the position and density of the defect states on the NCs. In blinking, the single NC shows clear on- and off-states. However, when such clear on- and offstates not observable, but instead, transition between several intensity levels is observed, the 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

Page 4 of 23

process is termed as flickering.18-20 It is found that excess trapped charge is one of the main reasons for PL-flickering in the NCs.18, 19 Recent studies on hybrid halide perovskites have shown, however, PL blinking is not restricted to particles in nanoscale dimension, but it can also be observed in larger size crystals (up to m).20-23 This finding is quite interesting as large size conventional semiconductors do not exhibit this kind of behavior. This interesting PL blinking is attributed to fluctuating nonradiative relaxation due to some metastable but efficient defect states.20-23 The all-inorganic perovskite NCs have emerged in recent years as highly promising material for photovoltaic, optoelectronic and photonic applications because of their enticing optical properties such as broad absorption, narrow tunable emission, high PL quantum yield (QY), large multi-photon absorption cross-section, high charge carrier mobility and tolerance to intrinsic defects.24-34 Despite the absence of deep trap states,34-37 recent reports show that CsPbX3 (X=Br, I) NCs are not completely free from charge trapping centers and are vulnerable towards photocharging, forming trions and dark excitons.30, 31, 38-41 Trion decay in these NCs is found to be dominated by nonradiative Auger recombination.40, 41 The mechanism of shallow trap-mediated nonradiative deactivation of the carriers in these NCs is not well understood at the moment. One of the approaches to probe the fate of the charge carriers of a semiconductor is by studying the PL properties in the single NC level. This is because interaction of the charge carriers with any internal or external defects leads to fluctuation in PL with time. Limited literature available on single particle PL behavior of CsPbX3 NCs reveals a number of patterns of PL fluctuation rather than a unifying trend.16, 30, 32, 33, 40, 42-44 Moreover, these studies do not present a consistent picture of the mechanism of observed PL fluctuation.16,

30, 39, 40, 42, 43

Some reports 4

ACS Paragon Plus Environment

Page 5 of 23 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

attribute blinking in these materials to random charging/ discharging.30,

39, 40

Another report

considers NBC recombination as the major reason for PL intermittency, where trion plays a minor role.16 Detailed information on the nature of the trions, reasons for the complex nature of the PL kinetics (of the bright state) and fluorescence lifetime-intensity distribution (FLID) are lacking. Considering these factors, in this work we attempt to provide a comprehensive picture of the charge carrier dynamics in single CsPbBr3 NC taking into consideration all possible radiative and nonradiative recombination pathways. CsPbBr3 NCs were synthesized following a recent room temperature anti-solvent precipitation method,45 the details of which are provided in ‘Methods’ section. Sample preparation for various measurements and instrumentation details are also provided in this section. Figure 1a shows transmission electron microscopy (TEM) images of the CsPbBr3 nanocubes indicating an average size of 12  1 nm. As evident from Figure 1b, first absorption onset and PL peak of these NCs appear at 505 nm and 512 nm, respectively. Confocal PL microscopic image of a film containing well-separated NCs are shown in Figure 1c. Individual NCs for PL microscopy measurements were excited with a 405 nm pulsed picosecond diode laser and PL signals were detected through time-tagged-time-resolved mode. Measurements were performed on nearly 60 individual single particles of as-synthesized and ethanethiol-treated NCs with an excitation intensity corresponding to N = 0.09, where N is the average number of photons absorbed by a NC after each pulse.

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

Page 6 of 23

Figure 1. (a) TEM and high resolution TEM images of CsPbBr3 NCs, (b) optical absorption and PL spectra of the NCs dispersed in toluene and (c) PL intensity image of a very low density NC film obtained using confocal microscope. Literature reports show that temporal fluctuations of PL intensity of single CsPbBr3 NCs can be of one of two following types; 18, 32, 33, 40, 42, 44 ‘PL blinking’, when the NCs undergo rapid transition between high intensity ON and low intensity OFF state32, 33, 40, 44 and ‘PL flickering’, when the NCs do not show clear ONOFF transitions, but exhibit a PL trajectory with continuous distribution of intensity as a function of time.18, 42, 44 In this work, we find that the as-synthesized NCs show both types of PL behavior; 30% of the studied NCs show blinking (Figure 2 and SI1) for which the PL intensity time-trace exhibits two distinct intensity regions corresponding to ON and OFF states and some intermediate intensity GREY states as well. The OFF states are not completely non-emissive for these NCs as the intensity corresponding to this state is quite above the background level (red line in Figure 2a). Blinking pattern of this nature is reported previously for CsPbBr3, as well as for CdSe/CdS quantum dot systems.5, 40, 43 Time-resolved PL (TRPL) intensity profiles corresponding to different intensity regions of the blinking time-trace (marked by R1, R2, R3 and R4) are shown in Figure 2b (Figure SI2). The highest intensity region (R1) shows a single long lifetime component of 8.90.8 ns. The other regions (R2, R3 and R4) are characterized by a biexponential decay, except in 40% cases, when lowest intensity OFF state (R4) 6 ACS Paragon Plus Environment

Page 7 of 23 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

shows a single lifetime component of 1.3 0.3 ns. Table 1 summarizes the individual lifetime components and their relative amplitudes for different intensity regions, averaged over all studied NCs which exhibit this blinking pattern. The data shows that each intensity level (R2/R3/R4) except R1 is associated with a long and a short lifetime component, representing two different recombination processes. Since multiple exciton formation in a NC is ruled out under our experimental condition and trions are short-lived compared to the single excitons, the long lifetime component (8.90.8 ns) associated with the highest intensity state is assigned to the single exciton and the shorter one to trion recombination. The FLID (Figure 2c) appears to be quite interesting as it is neither pure linear nor pure curvature; instead, it carries a combined signature of both. The dashed lines in Figure 2c present a guide to the eye. A linear FLID implies a near unity ratio of the radiative rates of two different intensity regions (defined hereafter as intensity-lifetime scaling (), SI for details) and only the nonradiative rates of the two levels differ. This kind of nature is attributed to NBC recombination without charging the NC.14, 15 A FLID with an  value of 2.0 suggests the involvement of trion recombination in the lower intensity states (as trion recombination rate is twice faster than that of neutral exciton).18, 30 The curved nature of our FLID suggests both NBC and trion recombinations contribute to the blinking of single CsPbBr3 NCs. Intensity-lifetime scaling of two small regions of the ON and OFF states in the blinking trace (Figure SI3) yields an  value of 1.22 confirming the contribution from both processes and also the fact that the short lifetime component (2) of the different intensity levels is due to trion recombination (influenced by the NBC recombination).

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

Page 8 of 23

Figure 2. (a) PL intensity time trace of a blinking NC with a binning time of 10 ms. (b) PL decay dynamics of different intensity levels marked (R1, R2, R3, R4) and (c) FLID with false color representation of single CsPbBr3 NC when N = 0.09.

Table 1. Average lifetime components and their relative amplitudes corresponding to different intensity regions of the blinking time-trace. Region R1 R2 R3 R4

1 (1)

8.9  0.8 9.2  0.4 (0.43) 8.1  0.6 (0.28) 6.7  0.6 (0.13)

2 (2) 1.6  0.4 (0.57) 1.3  0.4 (0.72) 1.0  0.2 (0.87) 1.3  0.3

8 ACS Paragon Plus Environment

Page 9 of 23 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

Even though the measurements have been carried out on single isolated NCs (Figure 1c), considering the fact that most of the intensity regions contain two lifetime components, it may appear that the observed blinking pattern is not due to single NCs, but arises from a cluster of NCs. This possibility however is ruled out considering the following: Let us assume the cluster to consist of two NCs. (a) If the NCs were somewhat different, then the decay profile for the highest intensity level should have been biexponential. However, this is not the case. (b) If the NCs were similar then the decay time constants would have been very similar with an -value of