Amplified Spontaneous Emission Threshold Reduction and Operational Stability Improvement in CsPbBr3 Nanocrystals Films by Hydrophobic Functionalization of the Substrate

The use of lead halide perovskites in optoelectronic and photonic devices is mainly limited by insufficient long-term stability of these materials. This issue is receiving growing attention, mainly owing to the operational stability improvement of lead halide perosvkites solar cells. On the contrary, fewer efforts are devoted to the stability improvement of light amplification and lasing. In this report we demonstrate that a simple hydrophobic functionalization of the substrates with hexamethyldisilazane (HMDS) allows to strongly improve the Amplified Spontaneous Emission (ASE) properties of drop cast CsPbBr3 nanocrystal (NC) thin films. In particular we observe an ASE threshold decrease down to 45% of the value without treatment, an optical gain increase of up to 1.5 times and an ASE operational stability increase of up to 14 times. These results are ascribed to a closer NC packing in the films on HMDS treated substrate, allowing an improved energy transfer towards the larger NCs within the NC ensemble, and to the reduction of the film interaction with moisture. Our results propose hydrophobic functionalization of the substrates as an easy approach to lower the ASE and lasing thresholds, while simultaneously increasing the active material stability.

Lead (II)-oleate (0.5 M) 4.6066 g of lead (II) acetate trihydrate (12 mmol, 1 eq) and 7.6 mL of oleic acid (24 mmol, 2 eq) were evacuated in a three-neck flask along with 16.4 mL of ODE at room temperature until the first gas evolution subsides and then further evacuated at 25-120 °C for 1 hour.

Isolation
To the crude solution (64 mL) 128 mL of ethyl acetate were added and the NC s were precipitated by centrifugation at 29500 xg (g is the earth gravitation constant) for 10 minutes.

Washing
The NCs were dispersed in 20 mL of toluene and precipitated with 40 mL of ethyl acetate and centrifugation at 29500 xg for 1 minute. During the second and third purification step solvents were reduced by factor 2 for each step. After the last precipitation the NCs were dispersed in toluene and centrifuged once more at 29500 xg for 10 min resulting in a 5 mg/mL colloidal solution.

Isolation
The crude solution was centrifuged at 29500 xg for 10 min the precipitate was collected and the supernatant discarded.

Washing
The NCs were dispersed in 20 mL of toluene and precipitated with 40 mL of ethyl acetate and centrifugation at 29500 xg for 1 minute. During the second and third purification step solvents were reduced by factor 2 for each step. After the last precipitation the NCs were dispersed in toluene and centrifuged once more at 29500 xg for 10 min resulting in a 8.6 mg/mL colloidal solution.

Substrates preparation
As a first step the glass substrates were sonicated in soap water, rinsed with deionized water, and dried by a compressed air jet (this procedure was repeated twice). Then they were sonicated in ethanol and dried with air jet then sonicated in acetone and dried with air jet.
The substrates then were either sonicated in hexane and dried with air jet before the films were deposited by drop casting 10 µL of the NC colloidal solution on substrates (films NC1 and NC2), or functionalized by drop casting 30µL of HMDS onto the glass substrate and letting it dry. The adsorbed HMDS was then annealed at 150 °C for 30 minutes and the films cooled back to room temperature during 1 hour before drop casting 10 µL of the same NC colloidal solution on the substrates (films NC1HMDS and NC2HMDS).

Substrates characterization
The wettability of the substrates have been investigated by water contact angle measurements. The equilibrium contact angle, averaged over 10 different distilled water droplets. is 60.7° for the untreated substrate, evidencing a wetting behavior. The HMDS functionalized substrates showed instead an average equilibrium contact angle of 89.0°, and are thus hydrophobic. Figure S1: TEM image of the NC1 (a) and NC2 (c) nanocrystals and corresponding histograms of the size distribution (b and d). The continuous line is the best fit curve with a Gaussian distribution.

S5
Absorption and photoluminescence spectra Figure S2: Absorbance and PL spectra of NC1 and NC2 in toluene solution. The spectra are normalized to 1 at the PL peak wavelength and at the exciton absorption peak for clarity. Figure S4. a): Excitation density dependence of the ASE peak intensity of the NC2 (black dots) and of the NC2HMDS (blue dots) samples. The lines are guide for the eyes. Inset: The same data plotted with a linear intensity scale, evidencing the strongly different ASE increase in the two samples. b): ASE peak intensity decrease during continuous laser pumping at an excitation density of 6.4 mJcm -2 in the NC2 sample (black dots) and of 2.8 mJcm -2 in the NC2HMDS sample (blue dots). The initial fast degradation is due to the ASE progressive quenching with pumping time while at longer times a slower degradation is observe, due to the spontaneous emission quenching. Simulated excitation density dependence of the ASE intensity