Effect of Passivation Layer on the Thin Film Perovskite Random Lasers

Novel functionalities of disorder-induced scattering effect in random lasers, attributed to low spatial coherence, draw remarkable attention in high-contrast to superior quality speckle-free imaging applications. This paper demonstrates perovskite-polystyrene (PS)-based random lasing action with robust optical performance at room temperature. Optical characterizations are carried out upon perovskite thin films addition with polystyrene of different mixing concentrations (wt.%). A low threshold lasing operation is achieved with an increasing concentration of polystyrene, accompanying a wavy surface texture with high surface roughness. The rough surface dominating multiple scattering effects leads to enhanced feedback efficiency. Moreover, this study also elucidates efficient fabrication process steps for the development of high quality and durable PS-based random lasers. With the advantages of reduced coherent artifacts and low spatial coherence, speckle free projection images of the USAF (U. S. Air Force MIL-STD-150A standard of 1951) resolution test chart are shown for different PS-based random lasers.


Introduction
Recent technological advancements in the research of lead halide perovskites have drawn remarkable attention in the context of optoelectronics device applications. Hybrid metal halide perovskites such as MAPbX 3 (MA-methyl ammonium and X-Cl, Br, or I) manifest as direct bandgap materials with large absorption coefficient, slow Auger recombination rate, low defect densities, and high optical gain. Furthermore, solvent engineered metal halide perovskites are deemed to be novel and extended researches to photovoltaics, LEDs, and lasers [1][2][3][4][5][6][7][8][9]. As reported, organic-inorganic lead halide perovskite-based laser devices are achieved by employing nanowires and whispering gallery modes for feedback oscillation [10][11][12]. Some other researches demonstrated resonators in tapered fiber design and self-organized nanorods in a microcrystalline network. Patterned and flattened perovskite thin films incorporated into distributed feedback (DFB) lasers, vertical cavity surface emitting lasers, photonic crystal lasers, and optical cavities were also reported. All such laser devices need to follow proper nano fabrication and controlled growth processes [13][14][15][16][17]. Thanks to the high optical gain in MAPbX 3 materials, only a thin film perovskite with proper scattering centers is able to achieve random lasing action without a complicated laser cavity fabrication process [18][19][20][21][22][23]. Also, the hybrid perovskite materials with Br content are suitable for high quality, and shape controlled single crystal formation. In addition, the solution processed organometal halide perovskite thin film of MAPbBr 3 is cost effective and shows faster crystallization with well interconnected grain boundaries [24,25].
in the second stage of spin coating process. At final process step, the sample was annealed at 100 • C for 30 min to vaporize the residual solvents as well as to transit from the intermediate solvate phase to perovskite thin film. Soon after the formation of perovskite thin film, different concentrations of polystyrene (wt.%) were deposited by the spin coating process with a rotating speed of 1000 rpm for 10 s. The respective polystyrene thin films of 10, 20, and 30 wt.% concentrations were formed after annealing it with 100 • C for 5 min.
Materials 2020, 13, x FOR PEER REVIEW  3 of 11 intermediate solvate phase to perovskite thin film. Soon after the formation of perovskite thin film, different concentrations of polystyrene (wt.%) were deposited by the spin coating process with a rotating speed of 1000 rpm for 10 s. The respective polystyrene thin films of 10, 20, and 30 wt.% concentrations were formed after annealing it with 100 °C for 5 min. The random scattering effect is provided by the microscale rugged surface morphology of perovskite films.

Micro-Photoluminescence Measurement Setup (μ-PL)
The experimental setup for micro-photoluminescence (μ-PL) measurement is shown in Figure  2. The optical pumping system applies Nd: YVO4, 355 nm pulse laser as a pumping source operated under 1 kHz repetition rate and 0.5 ns pulse width. A 0.55 numerical aperture (NA) microscopic objective lens is placed between the sample and the laser to collect the light reflected from the sample surface and transmit the light into a charge-coupled device (CCD); thus, a clear view of the sample surface can be seen. The experiments were carried out under room temperature. The excitation area dependent photoluminescence (PL) signal as a function of pumping energy density was received by a UV optical fiber of 600 μm core normal to the sample surface and fed into a nitrogen-cooled spectrometer (Jobin-Yvon IHR 320 Spectrometer, Kyoto, Japan) having the spectral resolution about 0.2 nm, through a charge-coupled device.

Micro-Photoluminescence Measurement Setup (µ-PL)
The experimental setup for micro-photoluminescence (µ-PL) measurement is shown in Figure 2. The optical pumping system applies Nd: YVO 4 , 355 nm pulse laser as a pumping source operated under 1 kHz repetition rate and 0.5 ns pulse width. A 0.55 numerical aperture (NA) microscopic objective lens is placed between the sample and the laser to collect the light reflected from the sample surface and transmit the light into a charge-coupled device (CCD); thus, a clear view of the sample surface can be seen. The experiments were carried out under room temperature. The excitation area dependent photoluminescence (PL) signal as a function of pumping energy density was received by a UV optical fiber of 600 µm core normal to the sample surface and fed into a nitrogen-cooled spectrometer (Jobin-Yvon IHR 320 Spectrometer, Kyoto, Japan) having the spectral resolution about 0.2 nm, through a charge-coupled device.  The random scattering effect is provided by the microscale rugged surface morphology of perovskite films.

Micro-Photoluminescence Measurement Setup (μ-PL)
The experimental setup for micro-photoluminescence (μ-PL) measurement is shown in Figure  2. The optical pumping system applies Nd: YVO4, 355 nm pulse laser as a pumping source operated under 1 kHz repetition rate and 0.5 ns pulse width. A 0.55 numerical aperture (NA) microscopic objective lens is placed between the sample and the laser to collect the light reflected from the sample surface and transmit the light into a charge-coupled device (CCD); thus, a clear view of the sample surface can be seen. The experiments were carried out under room temperature. The excitation area dependent photoluminescence (PL) signal as a function of pumping energy density was received by a UV optical fiber of 600 μm core normal to the sample surface and fed into a nitrogen-cooled spectrometer (Jobin-Yvon IHR 320 Spectrometer, Kyoto, Japan) having the spectral resolution about 0.2 nm, through a charge-coupled device.

Experimental Setup for USAF Resolution Test Chart Imaging
The experimental setup for the imaging USAF resolution test chart is shown in Figure 3. The Nd: YVO 4 , 355 nm pulse laser is applied as the pumping source onto the random laser sample. The excited random laser light emission is collected by the objective lens of N.A. of 0.45, and magnification of 20×. After passing through a diffuser, the laser is projecting onto an USAF resolution test chart. The transmitted light from the USAF test chart is then focused by an objective lens of N.A. of 0.7 with a magnification of 50×, thereafter imaged by a CCD camera.

Experimental Setup for USAF Resolution Test Chart Imaging
The experimental setup for the imaging USAF resolution test chart is shown in Figure 3. The Nd: YVO4, 355 nm pulse laser is applied as the pumping source onto the random laser sample. The excited random laser light emission is collected by the objective lens of N.A. of 0.45, and magnification of 20×. After passing through a diffuser, the laser is projecting onto an USAF resolution test chart. The transmitted light from the USAF test chart is then focused by an objective lens of N.A. of 0.7 with a magnification of 50×, thereafter imaged by a CCD camera.

Surface Texture Analysis
Planar view scanning electron microscope (SEM) images and atomic force microscope (AFM) surface morphologies of different samples are shown in Figure 4a,b, respectively. The disordered rugged surface morphology by randomly distributed perovskite grains is found over the substrate.

Surface Texture Analysis
Planar view scanning electron microscope (SEM) images and atomic force microscope (AFM) surface morphologies of different samples are shown in Figure 4a,b, respectively. The disordered rugged surface morphology by randomly distributed perovskite grains is found over the substrate. Planar views of add-on polystyrene thin films with a concentration varied from 10 to 30 wt.% show a hard-sticky film with rugged inhomogeneous wavy surface texture with incremental surface roughness.    Figure 5 displays the emission spectra of perovskite random lasers as a function of pumping energy density, and their corresponding L-L curves together with FWHM of emission spectra are shown in Figure 6. The diameter of the pumping spot is approximately 83  5 m. In principle, the random lasing action is achieved by the dominant multiple scattering effect. In this study, the dominant multiple scattering effect is provided by the microscale rugged surface morphology induced by inhomogeneous crystallographic perovskite grains as shown in Figure 1. With the increase of pumping energy density, the "s" shape of the L-L curve is observed in perovskitepolystyrene (wt.%) random lasers, accompanying the abrupt change of spectra linewidth at the measured threshold energy density. As shown in Figure 5a, the emission spectra may refer to a photoluminescent (PL) mode and a random lasing (RL) mode, coexisting in the light emission performance of the perovskite thin films. The central wavelengths of the PL and RL modes are at around 535 nm and 546 nm, respectively. With an increasement of the pumping powers, the full width at half maximum (FWHM) of the PL emission peak is almost invariant at around 24 nm. In terms of the RL mode, the band width may be reduced to about 6.17 nm at the maximum pumping energy density of 5.5 mJ/cm 2 . The extracted peak width at different pumping densities is presented in Figure 6a. Furthermore, multiple emission peaks can be observed in the RL mode, which may be one of the significant features of the random lasing action [39]. The measured intensities of each emission peaks may change along with the pumping power variation. Regarding to the lasing performance of the bare perovskite thin films, the L-L curve does not show a clear "s" shape in the log-log scale, which indicates that the perovskite sample only reaches the amplified spontaneous emission (ASE) performance with a threshold of 3.1 mJ/cm 2 . Further, a clear inference can be drawn from perovskite ASE spectral response that, it is evident of less spikes and wide spectral linewidth due to negligible optical feedback in comparison to other perovskite-polystyrene random lasers. This is a clear signature of diffusive light propagation due to long scattering mean free path. The emission spectra of perovskite-polystyrene random lasers with 10, 20, and 30 wt.% concentrations are shown in Figure 5b-d. The distinct multiple lasing peaks of perovskite-polystyrene random lasers are observed at varied pumping power comparing to the perovskite sample shown in Figure 5a. This indicates strong photon localization effect induced by the rugged surface morphologies with wavy surface texture. Different mode spacing could be a result of a deformation in cavity length due to external stress acquired by the wavy polystyrene surface texture [40,41]. As shown in Figure 5, redshifted lasing peaks with respect to photoluminescence (PL) peak at about 535 nm are spotted for different random lasers, which results from less band tail absorption loss, when the gain-loss competition is considered [42]. Thus, the addition of different concentrations of polystyrene as a passivation layer improves the disorder strength, and photon lifetime due to multireflection processes at the interfaces. This turns out to decrease the lasing threshold. As a result of this, the lowest threshold energy density of 2.4 mJ/cm 2 is achieved in perovskite-polystyrene 30 wt.% based  Figure 5 displays the emission spectra of perovskite random lasers as a function of pumping energy density, and their corresponding L-L curves together with FWHM of emission spectra are shown in Figure 6. The diameter of the pumping spot is approximately 83 ± 5 µm. In principle, the random lasing action is achieved by the dominant multiple scattering effect. In this study, the dominant multiple scattering effect is provided by the microscale rugged surface morphology induced by inhomogeneous crystallographic perovskite grains as shown in Figure 1. With the increase of pumping energy density, the "s" shape of the L-L curve is observed in perovskite-polystyrene (wt.%) random lasers, accompanying the abrupt change of spectra linewidth at the measured threshold energy density. As shown in Figure 5a, the emission spectra may refer to a photoluminescent (PL) mode and a random lasing (RL) mode, coexisting in the light emission performance of the perovskite thin films. The central wavelengths of the PL and RL modes are at around 535 nm and 546 nm, respectively. With an increasement of the pumping powers, the full width at half maximum (FWHM) of the PL emission peak is almost invariant at around 24 nm. In terms of the RL mode, the band width may be reduced to about 6.17 nm at the maximum pumping energy density of 5.5 mJ/cm 2 . The extracted peak width at different pumping densities is presented in Figure 6a. Furthermore, multiple emission peaks can be observed in the RL mode, which may be one of the significant features of the random lasing action [39]. The measured intensities of each emission peaks may change along with the pumping power variation. Regarding to the lasing performance of the bare perovskite thin films, the L-L curve does not show a clear "s" shape in the log-log scale, which indicates that the perovskite sample only reaches the amplified spontaneous emission (ASE) performance with a threshold of 3.1 mJ/cm 2 . Further, a clear inference can be drawn from perovskite ASE spectral response that, it is evident of less spikes and wide spectral linewidth due to negligible optical feedback in comparison to other perovskite-polystyrene random lasers. This is a clear signature of diffusive light propagation due to long scattering mean free path. The emission spectra of perovskite-polystyrene random lasers with 10, 20, and 30 wt.% concentrations are shown in Figure 5b-d. The distinct multiple lasing peaks of perovskite-polystyrene random lasers are observed at varied pumping power comparing to the perovskite sample shown in Figure 5a. This indicates strong photon localization effect induced by the rugged surface morphologies with wavy surface texture. Different mode spacing could be a result of a deformation in cavity length due to external stress acquired by the wavy polystyrene surface texture [40,41]. As shown in Figure 5, red-shifted lasing peaks with respect to photoluminescence (PL) peak at about 535 nm are spotted for different random lasers, which results from less band tail absorption loss, when the gain-loss competition is considered [42]. Thus, the addition of different concentrations of polystyrene as a passivation layer improves the disorder strength, and photon lifetime due to multireflection processes at the interfaces. This turns out to decrease the lasing threshold. As a result of this, the lowest threshold energy density of 2.4 mJ/cm 2 is achieved in perovskite-polystyrene 30 wt.% based random laser as shown in Figure 6d. Similarly, Figure 6b-d depict improved lasing characteristics with lowering down the threshold energy density values as 2.9 mJ/cm 2 , and 2.6 mJ/cm 2 for perovskite-polystyrene 10 wt.%, and perovskite-polystyrene 20 wt.% concentration-based random lasers respectively.

Lasing Performance Analysis
Materials 2020, 13, x FOR PEER REVIEW 6 of 11 random laser as shown in Figure 6d. Similarly, Figure 6b-d depict improved lasing characteristics with lowering down the threshold energy density values as 2.9 mJ/cm 2 , and 2.6 mJ/cm 2 for perovskitepolystyrene 10 wt.%, and perovskite-polystyrene 20 wt.% concentration-based random lasers respectively.

Optical Characteristics of Random Lasers and Speckle Free Imaging
To study the threshold characteristics of different random lasers, it is observed that threshold energy density decreases upon increase in excitation area. This can be comprehended by Raith and Apalkov statistical analysis [43][44][45][46] of threshold behavior using power-law.
where E th is the threshold energy density, A is the excitation area, A 0 is the two-dimensional area acquired by quasi mode, G is disorder strength of the medium, and λ depends on degree of disordered correlation. For simplicity, the coefficient λ → 1 is set as the first degree of disordered correlation.
Thus, Equation (1) reduces to E th ∝ A A 0 −1 G which facilitates the comparison of different random lasing systems. Figure 7a shows the excitation area dependent threshold variation, where the experimental data (dots) is fitted by theoretical calculation (lines). By using power-law estimation, the inverse of disorder strength (1/G) is fitted as 1.17, 1.13, and 0.93 for perovskite-polystyrene random laser with polystyrene concentration of 10, 20, and 30 wt.% concentrations, respectively. It is well evident that disorder strength (G) increases with an increase in polystyrene concentration. As a result, it is clear to testify that under the large excitation area, the increase of polystyrene concentrations exhibits higher light-matter interaction efficiency and a dominant multiple scattering effect. This is due to the presence of a wavy surface texture complemented by rugged inhomogeneous surface morphology. Figure 7b shows the image of the USAF (U. S. Air Force MIL-STD-150A standard of 1951) resolution test chart [47,48] under illumination of different light sources. It is well evident that the full field image projection of the USAF test chart fails under illumination of Nd: YAG coherent light source. The intense illumination source with high spatial coherence generates coherent image artifacts, hence leads to speckle or corrupt image formation. To mitigate such effects and to improve image quality, perovskite-polystyrene (wt.%) based random lasers with low spatial coherence are used to produce speckle free, full-field image projections of USAF resolution test chart. Results from the contrast to noise ratio, verify the improved image quality using different perovskite-polystyrene (wt.%) based random lasers. To evaluate the image quality numerically, the contrast to noise ratio (CNR) is applied by the calculation of , where <I f > and <I b > are average of foreground and background intensities, σ f and σ b are the standard deviations of foreground and background intensities. It can be clearly seen that the perovskite-polystyrene random laser with polystyrene concentration of 30 wt.% concentration demonstrates the best CNR values. Although the perovskite ASE manifests a relatively low coherent light source, the low spectra intensity more or less degrades the imaging quality in comparison to the random laser light source. Figure 7c demonstrates the durability test for different perovskite samples. It is evident that, with a 120 h experiment under ambient conditions, the random lasers show degradation and a drop in photoluminescence intensity. However, the higher slope of the exponential curves manifests an increased lifetime upon the addition of concentrations of polystyrene (wt.%) as a passivation layer on the top perovskite thin film. Upon addition of polystyrene (wt.%), it provides wavy surface texture with an increased surface roughness that induces dominant multiple scattering effects which accounts for low threshold operation. An increased number of scattering centers that exhibit higher radiation efficiency incorporated into a longer lifetime prevails in perovskite-polystyrene (wt.%) random lasers.

Optical Characteristics of Random Lasers and Speckle Free Imaging
To study the threshold characteristics of different random lasers, it is observed that threshold energy density decreases upon increase in excitation area. This can be comprehended by Raith and Apalkov statistical analysis [43][44][45][46] of threshold behavior using power-law.  increased lifetime upon the addition of concentrations of polystyrene (wt.%) as a passivation layer on the top perovskite thin film. Upon addition of polystyrene (wt.%), it provides wavy surface texture with an increased surface roughness that induces dominant multiple scattering effects which accounts for low threshold operation. An increased number of scattering centers that exhibit higher radiation efficiency incorporated into a longer lifetime prevails in perovskite-polystyrene (wt.%) random lasers.

Conclusions
In this study, we have demonstrated solvent-engineered efficient fabrication process steps for the development of high quality and durable perovskite-polystyrene (PS)-based random lasers with robust optical performance at room temperature. The addition of different concentrations of polystyrene (wt.%) on perovskite thin films exhibit a long lifetime and low threshold random lasing operation as a reason of higher surface roughness induced dominant multiple scattering effect. In consequence of low spatial coherence and reduced coherent image artifacts, perovskite-polystyrene (PS) based random lasers manifest full-field speckle free image projections of USAF resolution test chart. From the comparative analyses among all random lasers, the perovskite-polystyrene (30 wt.%) based random laser demonstrates the lowest threshold operation with the best CNR values in speckle free imaging. With an advantage of speckle free imaging, perovskite-polystyrene (PS) based random lasers provides a greater potential application to optical coherence tomography, the formation of the colored image, and image processing, etc.