Study of Construction and Performance on Photoelectric Devices of Cs–Pb–Br Perovskite Quantum Dot

White LEDs were encapsulated using Cs4PbBr6 quantum dots and Gd2O3:Eu red phosphor as lamp powder. Under the excitation of a GaN chip, the color coordinates of the W-LED were (0.33, 0.34), and the color temperature was 5752 K, which is close to the color coordinate and color temperature range of standard sunlight. The electric current stability was excellent with an increase in the electric current, voltage, and luminescence intensity of the quantum dots and phosphors by more than 10 times. However, the stability of the quantum dots was slightly insufficient over long working periods. The photocatalytic devices were constructed using TiO2, CsPbBr3, and NiO as an electron transport layer, light absorption layer, and catalyst, respectively. The Cs–Pb–Br-based perovskite quantum dot photocatalytic devices were constructed using a two-step spin coating method, one-step spin coating method, and full PLD technology. In order to improve the water stability of the device, a hydrophobic carbon paste and carbon film were selected as the hole transport layer. The TiO2 layer and perovskite layer with different thicknesses and film forming qualities were obtained by changing the spin coating speed. The influence of the spin coating speed on the device’s performance was explored through SEM and a J–V curve to find the best spin coating process. The device constructed by the two-step spin coating method had a higher current density but no obvious increase in the current density under light, while the other two methods could obtain a more obvious light response, but the current density was very low.


Introduction
In the field of lighting display, LEDs, LCDs, and OLEDs are the main devices used presently. As a basic light-emitting device, LEDs have a wide range of application in household lighting [1][2][3][4], traffic signals, landscape decoration, and other fields. White LED devices usually adopt three packaging methods: a red, green, and blue chip combination; red, green, and blue chip activated phosphor; blue chip activated red, green, and blue phosphor. Compared with the other two packaging methods, the Blu-ray chip-excited red-green phosphor packaging method has the advantages of having a simple process, good stability, and low cost. However, as a typical green light type of quantum dots, the research on lead bromide-based perovskite quantum dots is less than that in the field of LED packaging.
The perovskite quantum dots with a Cs-Pb-Br structure have the advantages of small size, high quantum efficiency, wide color gamut, narrow half-peak width, and high carrier mobility. Compared with traditional phosphors, perovskite quantum dots are more energy saving and environmentally friendly and more suitable for the production of small-size or mini-LEDs. The stability of Cs 4 PbBr 6 perovskite quantum dots is better than that of CsPbBr 3 [5][6][7][8][9][10]. Therefore, Cs 4 PbBr 6 perovskite quantum dots have high application value in encapsulated white LED devices.
In this paper, Cs 4 PbBr 6 perovskite quantum dots, red phosphors, and LED chips were used to encapsulate white LED, and the performance of the devices was studied by EL spectrum and color coordinates. TiO 2 as the electron transport layer, CsPbBr 3 perovskite for the light absorption layer, carbon paste and carbon film for the holes transport layer, and an NiO catalyst layer were used. A photocatalytic device based on a multilayer thin-film solar cell structure was constructed using the solution spin coating method and pulsed laser deposition technology [11][12][13][14][15][16][17]. The effects of the spin coating rate on the TiO 2 layer and perovskite layer are discussed, while the thickness and film quality of the TiO 2 layer and perovskite layer on the performance of the device were analyzed.

Characterization
The morphology of the precursor and resultant products were collected via fieldemission scanning electron microscope (FE-SEM, QUANTA FEG 250, FEI Co., Waltham, MA, USA) with an acceleration voltage of 10 kV. The electrochemical performance of the samples was tested on an electrochemical workstation (Zahner/Zennium, Bavaria, Germany). The PLE and PL spectra were obtained using a Fluorescence Spectrophotometer (FP-6500, JASCO Co., Tokyo, Japan) at room temperature equipped with a Φ60 mm integrating sphere (ISF-513, JASCO, Tokyo, Japan) and a 150 W Xe-lamp as the excitation source. The optical performances for all samples were conducted under identical conditions with a slit breadth of 5 nm. The specimen was excited with a selected wavelength, and the intensity of the intended emission was recorded as a function of elapsed time after the excitation light was automatically cut-off using a shutter.

Synthesis of Cs 4 PbBr 6
Both 0.0213 g (0.1 mmol) CsBr and 0.0368 g (0.1 mmol) PbBr 2 were dissolved in 5 mL DMF solution, which contained 30 µL OAM and 0.5 mL OA. After ultrasonic dispersion and stirring for 15 min, the quantum dot precursor solution was obtained. The precursor solution 0.4 mL was injected into 5 mL of toluene solution, which was violently stirred continuously. A reaction occurred after 5 s, and the yellow-green colloid solution was formed and Cs 4 PbBr 6 quantum dots were obtained. The colloid solution was placed in a centrifuge tube and centrifuged at 12,000 RPM for 5 min. After being washed twice with toluene, the precipitation was dried in a vacuum oven at 60 • C for 24 h.

Synthesis of Gd 2 O 3 :Eu Red Phosphor
Accurate weights of Gd 2 O 3 and Eu 3+ phosphors according to the stoichiometric ratio of 1:0.05. Then, 2.85 mL Gd(NO 3 ) 3 solution and 0.75 mL Eu (NO 3 ) 3 solution were added to a 250 mL beaker, stirred for 10 min after adding a suitable amount of deionized water, adding suitable amount of urea and stirring for 10 min to ensure it fully dissolved. The mixed solution was heated up to 90 • C quickly and maintained for 2 h at 90 • C until the reaction was complete.
The precipitate was obtained by centrifugation when the reaction solution was cooled to room temperature. The precipitate was cleaned twice with distilled water and anhydrous ethanol and dried at 80 • C. The precursor powder was calcined at 800 • C with a heating rate of 5 • C/min and a holding time of 4 h. The atmosphere of the calcination was air. After calcination, Gd 2 O 3 : Eu 3+ phosphors were obtained.

Encapsulation of White LED
Bothe 0.005 g Cs 4 PbBr 6 powder and 0.035 g Gd 2 O 3 :Eu red phosphor were placed in an agate mortar. They were fully ground and mixed. Then, 0.4 g of silicone resin A glue was added to the stir and mixed, after which 0.05 g of silicone resin B glue was added to lay the mixed phosphor on the GaN chip. After the removal of bubbles, the solidification was accelerated by heating. Next, 0.005 g Cs 4 PbBr 6 powder and 0.035 g Gd 2 O 3 :Eu red phosphor were fully ground and mixed in an agate mortar, and 0.4 g of silicone resin A glue was added. After stirring and mixing, 0.05 g of silicone resin B glue was added. The mixed phosphor was tiled on the GaN chip and heated to accelerate its solidification after removing the bubbles.

Photoelectric Catalysis Devices Based on CsPbBr 3 by Two Step Spin Coating
Photocatalytic devices based on CsPbBr 3 are similar to multilayer solar cells or Q-LED devices. The glass substrate containing fluorine-doped tin oxide (FTO) was surface etched on and was cut into a 3 cm × 3 cm 2 , and then ultrasonic washed with acetone, methanol, and deionized water for 15 min. It was dried in N 2 atmosphere and then exposed to UV-ozone for 15 min [18,19].
Then, 0.5 mL titanate was dissolved in 5 mL ethanol and spin-coated for 50 s at 2000-4000 RPM twice, followed by annealing at 500 • C for 60 min to form the TiO 2 electron transport layer. Lead bromide was dissolved in anhydrous DMF (1 mmol/mL) and cesium bromide was dissolved in methanol (0.1 mmol/mL), which was spin-coated for 30 s at 2000-4000 RPM successively and then vacuum annealed at 300 • C for 30 min to form the perovskite layer. The carbon paste was scraped onto the perovskite layer by hand, and the graphite sheet was cut into 3 cm × 2.2 cm and pasted on the carbon paste. The carbon paste was annealed by vacuum at 120 • C for 60 min in order to solidify the film. The catalyst layer was deposited using the pulsed laser deposition (PLD) technique and annealed in vacuum at 200 • C for 120 min. Finally, the device was encapsulated with UV drops.

Photoelectric Catalysis Devices Based on CsPbBr 3 by One Step Spin Coating
Both 0.0213 g (0.1 mmol) CsBr and 0.0440 g (0.12 mmol) PbBr 2 were dissolved in 5 mL DMF that contained 30 µL OAM and 0.5 mL OA. The solution was stirred to dissolve all the drugs and the precursor solution was obtained after 15 min. The precursor solution (0.4 mL) was injected into a vigorously agitated toluene solution (5 mL). After reaction for 5 s, the colloidal solution was immediately centrifuged at 12,000 RPM for 5 min. The supernatant was discarded, the precipitation was dispersed in toluene for centrifugation, and CsPbBr 3 quantum dots were obtained.
After the FTO glass was treated, the titanate ethanol solution was spin-coated 50 s at 3000 RPM and repeated twice, and the TiO 2 electron transport layer was formed after annealing at 500 • C for 60 min. The toluene solution of CsPbBr 3 quantum dots was spincoated for 60 s at 3000 RPM and repeated twice. The solution was annealed in a vacuum at 100 • C for 60 min. The remaining steps were consistent with Section 2.4.

Photoelectric Catalysis Devices Prepared by PLD Technique
Low-pressure pulsed laser deposition (PLD) was used for the TIO 2 layer deposition. The TiO 2 layer was formed after annealing at 500 • C for 1 h with a pulse frequency of 30,000 at 5 Hz and 100 mJ.
The lead bromide target and cesium bromide target were also deposited at the frequency of 5 Hz, 100 mJ and energy of 20,000 pulses, and the perovskite layer was formed by vacuum annealing at 300 • C for 30 min. The NiO layer was grown on the nickel oxide target with 150 mJ energy and a 5 Hz pulse frequency of 10,000, and annealed in a vacuum at 200 • C for 120 min.

Results and Discussion
In order to study the application value of Cs 4 PbBr 6 quantum dots in lighting technology, we mixed Cs 4 PbBr 6 quantum dots and red Gd 2 O 3 :Eu phosphor as lamp powder, encapsulated them on GaN chip at 380 nm using silicone resin, and assembled simple white light-emitting diodes (W-LEDs). Figure 1a shows the EL spectrum of a W-LED. Three emission peaks are intuitively displayed in the spectrum. The emission peak at 380 nm corresponded to the emission peak of the chip, the emission peak at 511 nm was the emission peak of Cs 4 PbBr 6 quantum dots excited by the chip. The satellite peak at 612 nm was the emission peak of Gd 2 O 3 :Eu. The illustration in Figure 1a shows the emission photos of the W-LED when it was energized and non-energized. It can be seen that under the excitation of the blue chip in the energized state, the mixed light of green quantum dots and red fluorescent powder presents white light emission. target with 150 mJ energy and a 5 Hz pulse frequency of 10,000, and annealed in a vacuu at 200 °C for 120 min.

Results and Discussion
In order to study the application value of Cs4PbBr6 quantum dots in lighting techn ogy, we mixed Cs4PbBr6 quantum dots and red Gd2O3:Eu phosphor as lamp powder, e capsulated them on GaN chip at 380 nm using silicone resin, and assembled simple wh light-emitting diodes (W-LEDs). Figure 1a shows the EL spectrum of a W-LED. Three emission peaks are intuitive displayed in the spectrum. The emission peak at 380 nm corresponded to the emissi peak of the chip, the emission peak at 511 nm was the emission peak of Cs4PbBr6 quantu dots excited by the chip. The satellite peak at 612 nm was the emission peak of Gd2O3:E The illustration in Figure 1a shows the emission photos of the W-LED when it was en gized and non-energized. It can be seen that under the excitation of the blue chip in t energized state, the mixed light of green quantum dots and red fluorescent powder p sents white light emission. According to the emission of W-LEDs, their color coordinates are depicted, resp tively, in CIE chromaticity diagram. As shown in Figure 1b, the color coordinates Cs4PbBr6 quantum dots and Gd2O3:Eu red phosphor were (0.07, 0.68) and (0.65, 0.35), spectively. The color coordinates of the encapsulated W-LED were (0.33, 0.34), and color temperature was calculated to be 5752 K, which is close to the color temperatu range of standard sunlight (5200~5500 K).
In the practical application of the W-LED, its current fluctuated generally. For t reason, the EL curves of the lamp beads at 40 mA to 300 mA were tested, as shown Figure 2a. As the current increases from 40 mA to 300 mA, the voltage at both ends of t lamp bead's increases from 3.06 to 3.31 V, and the luminescence intensity of both quantu dots and phosphors increased more than 10 times. This indicates that the W-LED bas on Cs4PbBr6 had good current and voltage stability. In addition, under the current of V and 40 mA, the changes in the emission intensity of the W-LED under the energiz state for 8 h were tested as shown in Figure 2b. It can be seen that the luminescence inte sity of Cs4PbBr6 quantum dots gradually decreased with the extension of time. After 8 of aging, the luminescence intensity of the Cs4PbBr6 quantum dots only retained 65.4 of their initial value, which indicates that the stability of the quantum dots was sligh According to the emission of W-LEDs, their color coordinates are depicted, respectively, in CIE chromaticity diagram. As shown in Figure 1b, the color coordinates of Cs 4 PbBr 6 quantum dots and Gd 2 O 3 :Eu red phosphor were (0.07, 0.68) and (0.65, 0.35), respectively. The color coordinates of the encapsulated W-LED were (0.33, 0.34), and its color temperature was calculated to be 5752 K, which is close to the color temperature range of standard sunlight (5200~5500 K).
In the practical application of the W-LED, its current fluctuated generally. For this reason, the EL curves of the lamp beads at 40 mA to 300 mA were tested, as shown in Figure 2a. As the current increases from 40 mA to 300 mA, the voltage at both ends of the lamp bead's increases from 3.06 to 3.31 V, and the luminescence intensity of both quantum dots and phosphors increased more than 10 times. This indicates that the W-LED based on Cs 4 PbBr 6 had good current and voltage stability. In addition, under the current of 3.1 V and 40 mA, the changes in the emission intensity of the W-LED under the energized state for 8 h were tested as shown in Figure 2b. It can be seen that the luminescence intensity of Cs 4 PbBr 6 quantum dots gradually decreased with the extension of time. After 8 h of aging, the luminescence intensity of the Cs 4 PbBr 6 quantum dots only retained 65.43% of their initial value, which indicates that the stability of the quantum dots was slightly insufficient. This may have been caused by the lack quantum dots' lack of optical stability, which needs to be further optimized.
terials 2021, 14, x FOR PEER REVIEW 5 o insufficient. This may have been caused by the lack quantum dots' lack of optical stabi which needs to be further optimized. The long-term stability of the quantum dot-based LED was poorer than the Gd2O red phosphor LED, which was determined by the characteristics of the quantum d themselves. Perovskite materials will degrade under the influence of air, water, ther radiation, and light radiation and will then lose their original photoelectric propert Roda et al. found that O2 could combine with photo-generated electrons to cause quenching without defect guidance [20]. The influence of light on perovskite mater was mainly reflected in two aspects: one is that photogenerated electrons will prom the destruction of oxygen on perovskite, the other is the photodegradation under the j action of surface defects. When perovskite absorbs photons, it produces charge carri which may spread to the crystal surface. The defect acts as an electron trap on the cry surface and the trap's photocarriers, forming a local trap state. With the extension of i mination time, the density of the trap gradually increases, which further disrupts the c tal structure and generates an electric field, further intensifying the coupling between riers and lattice, ultimately leading to lattice distortion and structural degradation of p ovskite [21]. Gd2O3:Eu red phosphor is a kind of phosphor in essence, which is supe to the perovskite quantum dot in stability.
In order to explore the application potential of CsPbBr3 perovskite materials in p toelectric decomposition water, TiO2 and NiO were selected as the electron transport la (ETM) and catalytic materials, respectively. According to the principle of energy le matching, based on the determination of CsPbBr3 energy levels [22][23][24][25], the schematic agram of their energy level structure was drawn as shown in Figure 3a. The long-term stability of the quantum dot-based LED was poorer than the Gd 2 O 3 :Eu red phosphor LED, which was determined by the characteristics of the quantum dots themselves. Perovskite materials will degrade under the influence of air, water, thermal radiation, and light radiation and will then lose their original photoelectric properties. Roda et al. found that O 2 could combine with photo-generated electrons to cause PL quenching without defect guidance [20]. The influence of light on perovskite materials was mainly reflected in two aspects: one is that photogenerated electrons will promote the destruction of oxygen on perovskite, the other is the photodegradation under the joint action of surface defects. When perovskite absorbs photons, it produces charge carriers, which may spread to the crystal surface. The defect acts as an electron trap on the crystal surface and the trap's photocarriers, forming a local trap state. With the extension of illumination time, the density of the trap gradually increases, which further disrupts the crystal structure and generates an electric field, further intensifying the coupling between carriers and lattice, ultimately leading to lattice distortion and structural degradation of perovskite [21]. Gd 2 O 3 :Eu red phosphor is a kind of phosphor in essence, which is superior to the perovskite quantum dot in stability.
In order to explore the application potential of CsPbBr 3 perovskite materials in photoelectric decomposition water, TiO 2 and NiO were selected as the electron transport layer (ETM) and catalytic materials, respectively. According to the principle of energy level matching, based on the determination of CsPbBr 3 energy levels [22][23][24][25], the schematic diagram of their energy level structure was drawn as shown in Figure 3a.
Perovskite material degrades due to the hydrolytic reaction in contact with water, which will lead to device failure [26]. Hydrolysis reactions must be carried out in solution. Therefore, in order to improve the water stability of the device, a combination of carbon paste and carbon film was used as the hole transport layer (HTM). As a good conductor, carbon has a very strong ability for hole transport, and the dense carbon film was a hydrophobic material. Such a hole transport layer provided both waterproof protection and hole transmission.  Perovskite material degrades due to the hydrolytic reaction in contact with wat which will lead to device failure [26]. Hydrolysis reactions must be carried out in solutio Therefore, in order to improve the water stability of the device, a combination of carb paste and carbon film was used as the hole transport layer (HTM). As a good conduct carbon has a very strong ability for hole transport, and the dense carbon film was a h drophobic material. Such a hole transport layer provided both waterproof protection a hole transmission.
The design and manufacture of a photocatalytic device with a standard multilay thin-film solar cell structure is shown in Figure 3b. Using titanate as precursor to sp coating on the FTO glass substrate as electron transport layer, the antisolvent spin coati method was used to deposit perovskite on the top of TiO2 layer. Carbon paste was us to cover the surface of CsPbBr3 as the hole transport material layer, and a graphite co ductive sheet was used as the encapsulation and conducting material. Ultrathin nick oxide layer was deposited by PLD as the outermost layer of the catalytic device. Final the device was encapsulated with UV curable adhesive. Each layer should be as flat possible with minimal porosity.
As shown in Figure 3c, the photocatalytic device was completely immersed in t electrolyte to construct a photochemical cell with reference electrode and counter-el trode. As shown in Figure 3d, under sunlight (from backlight), photogenerated electro in CsPbBr3 were transferred by the TiO2 layer to the FTO for collection, and then flow through the external circuit to the opposite electrode, where the water-receiving electro on the surface of the opposite electrode are reduced to produce hydrogen. The holes m grated through the carbon layer to the NiO layer to produce oxygen by oxidizing the s rifice electrolyte (KOH solution) in the solution. Importantly, the NiO layer at the top n only acted as a physical barrier for electrolyte penetration into the active perovskite lay The design and manufacture of a photocatalytic device with a standard multilayer thin-film solar cell structure is shown in Figure 3b. Using titanate as precursor to spin coating on the FTO glass substrate as electron transport layer, the antisolvent spin coating method was used to deposit perovskite on the top of TiO 2 layer. Carbon paste was used to cover the surface of CsPbBr 3 as the hole transport material layer, and a graphite conductive sheet was used as the encapsulation and conducting material. Ultrathin nickel oxide layer was deposited by PLD as the outermost layer of the catalytic device. Finally, the device was encapsulated with UV curable adhesive. Each layer should be as flat as possible with minimal porosity.
As shown in Figure 3c, the photocatalytic device was completely immersed in the electrolyte to construct a photochemical cell with reference electrode and counter-electrode. As shown in Figure 3d, under sunlight (from backlight), photogenerated electrons in CsPbBr 3 were transferred by the TiO 2 layer to the FTO for collection, and then flowed through the external circuit to the opposite electrode, where the water-receiving electrons on the surface of the opposite electrode are reduced to produce hydrogen. The holes migrated through the carbon layer to the NiO layer to produce oxygen by oxidizing the sacrifice electrolyte (KOH solution) in the solution. Importantly, the NiO layer at the top not only acted as a physical barrier for electrolyte penetration into the active perovskite layer but also acted as an effective oxidation catalyst with generated cavities [27].
We explored the spin coating process of the TiO 2 layer at different spin coating speeds as shown in Figure 4. After annealing at 500 • C, dense TiO 2 was formed by spin-coating the titanate/ethanol solution at 2000, 3000, and 4000 RPM, respectively. As can be seen from the SEM image of cross-section in Figure 4, with the increase in the spin-coating rate, the thickness of the TiO 2 film decreased from 295 to 120 nm. At the same time, it can be seen from SEM surface scan that dense TiO 2 films were formed at the three rotational speeds, but the films formed at 2000 RPM and 4000 RPM were slightly rough. Due to the low rotational speed at 2000 RPM, agglomerated small particles existing in titanate/ethanol solution will still remain on the film after spinning. TiO 2 film remained rough after annealing. Because the surface of the FTO glass was rough, and the TiO 2 film formed at 4000 RPM had a small thickness that was not enough to make up for the uneven surface of the FTO; thus, a rough TiO 2 film formed. However, the film formed at 3000 RPM was relatively thick, and the high rotating speed can ensure that large particles will not remain, so a smoother, compact and flat TiO 2 film is formed. thickness of the TiO2 film decreased from 295 to 120 nm. At the same time, it can be se from SEM surface scan that dense TiO2 films were formed at the three rotational spee but the films formed at 2000 rpm and 4000 rpm were slightly rough. Due to the low ro tional speed at 2000 rpm, agglomerated small particles existing in titanate/ethanol so tion will still remain on the film after spinning. TiO2 film remained rough after anneali Because the surface of the FTO glass was rough, and the TiO2 film formed at 4000 rp had a small thickness that was not enough to make up for the uneven surface of the FT thus, a rough TiO2 film formed. However, the film formed at 3000 rpm was relativ thick, and the high rotating speed can ensure that large particles will not remain, s smoother, compact and flat TiO2 film is formed. Its J-V curve was obtained as shown in Figure 5. The current density of the devi with three spin coating speeds increased to a certain extent after illumination (10.99 14.30%, 21.80%, respectively with the increase in rotation speed. This was due to the den and flat TiO2 film, which obtained the highest current density and higher photocurr enhancement at 3000 RPM. In addition, we found that there was a certain current dens when the voltage was 0 V, which may be attributed to the transfer of electrons and ho in the perovskite layer under the irradiation of ambient light [18]. The current dens Its J-V curve was obtained as shown in Figure 5. The current density of the devices with three spin coating speeds increased to a certain extent after illumination (10.99%, 14.30%, 21.80%, respectively with the increase in rotation speed. This was due to the dense and flat TiO 2 film, which obtained the highest current density and higher photocurrent enhancement at 3000 RPM. In addition, we found that there was a certain current density when the voltage was 0 V, which may be attributed to the transfer of electrons and holes in the perovskite layer under the irradiation of ambient light [18]. The current density variation between 0.35 and 0.55 V was attributed to the REDOX potential of the NiO hydrate NiOOH. The spinning coating technology for the CsPbBr3 layer is further discussed belo The CsPbBr3 layer was prepared by a two-step spin coating process. The first step w spin-coating the PbBr2 DMF solution, and the second step was spin-coating the CsBr me anol solution. The CsPbBr3 layer was formed after annealing. The spin-coating speed PbBr2 and CsBr was adjusted at the same time to obtain CsPbBr3 layers with differ characteristics. Its performance characteristics were characterized by SEM images. shown in Figure 6, compared with TiO2 layer, the film forming quality of CsPbBr3 lay was relatively poor, and the porosity was higher. At 2000 rpm, the thickness reached 4 nm. The higher layer thickness made the surface of the CsPbBr3 layer uneven and full small pores. The thickness of the perovskite layer at 3000 rpm was approximately 238 n but the pore size of the perovskite layer is increased, although the number of pores reduced compared with that at 2000 rpm. Only when the rotating speed was increased 4000 rpm, a relatively smooth, dense, and low-porosity CsPbBr3 layer could be obtain and the thickness of the layer was 133 nm. In combination with the discussion on the T layer technology, we used a spinning speed of 3000 rpm to obtain the TiO2 layer, and spinning speed of 4000 rpm was used to obtain the CsPbBr3 layer. After scraping a The spinning coating technology for the CsPbBr 3 layer is further discussed below. The CsPbBr 3 layer was prepared by a two-step spin coating process. The first step was spin-coating the PbBr 2 DMF solution, and the second step was spin-coating the CsBr methanol solution. The CsPbBr 3 layer was formed after annealing. The spin-coating speed of PbBr 2 and CsBr was adjusted at the same time to obtain CsPbBr 3 layers with different characteristics. Its performance characteristics were characterized by SEM images. As shown in Figure 6, compared with TiO 2 layer, the film forming quality of CsPbBr 3 layer was relatively poor, and the porosity was higher. At 2000 RPM, the thickness reached 460 nm. The higher layer thickness made the surface of the CsPbBr 3 layer uneven and full of small pores. The thickness of the perovskite layer at 3000 RPM was approximately 238 nm, but the pore size of the perovskite layer is increased, although the number of pores is reduced compared with that at 2000 RPM. Only when the rotating speed was increased to 4000 RPM, a relatively smooth, dense, and low-porosity CsPbBr 3 layer could be obtained, and the thickness of the layer was 133 nm. In combination with the discussion on the TiO 2 layer technology, we used a spinning speed of 3000 RPM to obtain the TiO 2 layer, and a spinning speed of 4000 RPM was used to obtain the CsPbBr 3 layer. After scraping and coating the carbon paste and sticking the graphite sheet, the obtained device was scanned again for the cross-section as shown in Figure 6d. The thickness of TiO 2 layer and CsPbBr 3 layer obtained in this study were 179 and 143 nm, respectively. A digital photo of the encapsulated device is shown in Figure 6d, where the rectangular window with no glue in the middle is the test surface, and the upper window is the electrode clamp. The TiO2 layer was obtained by spin coating at 3000 rpm, and the perovskite lay was obtained at different spinning coating speeds. Using a 300 W xenon lamp as lig source for continuous illumination. Under continuous lighting conditions, the elect chemical workstation automatically measured. The frequency was measured six tim every 4 min. The J-V curves of the device are shown in Figure 7. The devices obtained the three spinning speeds all had higher current densities. Under light condition, the c rent density of the devices obtained at 2000, 3000, and 4000 rpm increased by 24.49 The TiO 2 layer was obtained by spin coating at 3000 RPM, and the perovskite layer was obtained at different spinning coating speeds. Using a 300 W xenon lamp as light source for continuous illumination. Under continuous lighting conditions, the electrochemical workstation automatically measured. The frequency was measured six times every 4 min.
The J-V curves of the device are shown in Figure 7. The devices obtained at the three spinning speeds all had higher current densities. Under light condition, the current density of the devices obtained at 2000, 3000, and 4000 RPM increased by 24.49%, 11.18%, and 38.27%, respectively. Because the layer of CsPbBr 3 was relatively flat, the current density and the increase in the current density at 4000 RPM were the highest. Due to the large pore area of the perovskite layer obtained at 3000 RPM, the current density of the J-V curve decreased after 0.55 V, which led to the leakage phenomenon. As shown in Figure 7d, the J-V curve of the photocatalytic device obtained by spin-coating at 4000 RPM was tested many times under illumination. The current density increased first and then stabilized. This indicates that the activation of the CsPbBr 3 layer took a certain time, and the best performance of the device could only be achieved after a period of illumination.
rials 2021, 14, x FOR PEER REVIEW 10 of technology will form a very dense film but will also form a layer of plasma glow betwe the layers, blocking the migration of electrons between layers, resulting in a significa increase in electrical resistance. In addition, a one-step method (spin-coated pre-prepared CsPbBr3 quantum dot t uene solution) was also attempted to replace the two-step method to prepare perovsk quantum dot layers. The J-V curve of the device obtained is shown in Figure 8a. Co pared with the two-step preparation process, the optical response of the device obtain by the one-step method was more obvious, and the current density under illuminati increased by 90.80%, which was much higher than the 38.27% of the two-step metho However, the maximum current density was only 2.09 mA/cm 2 , which was much low than 23.25 mA/cm 2 . This was because the spin-coating solution of the one-step coati method was the solution of the CsPbBr3 quantum dots. Perovskite quantum dots be This was similar to the device obtained by the one-step method. PLD devices also had an obvious optical response. When the light source was turned on, the current density increased to 118.58%, but its current density was only 0.036 mA/cm 2 . The higher current density may be related to the dense perovskite deposits. The use of PLD laser deposition technology will form a very dense film but will also form a layer of plasma glow between the layers, blocking the migration of electrons between layers, resulting in a significant increase in electrical resistance.
In addition, a one-step method (spin-coated pre-prepared CsPbBr 3 quantum dot toluene solution) was also attempted to replace the two-step method to prepare perovskite quantum dot layers. The J-V curve of the device obtained is shown in Figure 8a. Compared with the two-step preparation process, the optical response of the device obtained by the one-step method was more obvious, and the current density under illumination increased by 90.80%, which was much higher than the 38.27% of the two-step method. However, the maximum current density was only 2.09 mA/cm 2 , which was much lower than 23.25 mA/cm 2 . This was because the spin-coating solution of the one-step coating method was the solution of the CsPbBr 3 quantum dots. Perovskite quantum dots been formed, so the spin-coating quantum dots had a high content and dense film formation and, thus, had a high light response current density. In the two-step coating method, the reagents that can generate perovskite were rotated on glass substrate (FTO) two times. The formation of perovskite quantum dots requires imaging in the process of rotating coating. Due to the limited contact surface and uneven mixing of the two-step coating reagents, the quantum dots content of the perovskite layer that formed was low, and the film quality was inferior to that of the one-step coating method. It may be that the interlayer resistance was high because the spin-coated QDs w not annealed at high temperature. We also tried to use PLD deposition to form TiO2 a CsPbBr3 layers, and the J-V curve obtained is shown in Figure 8b. This was similar to devices obtained by the one-step method, the PLD devices also had obvious optical sponse. The current density increased to 118.58%, but the current density was only 0. mA/cm 2 . The higher rate of current density increase may be related to the dense perovs deposits. Very dense thin films can be formed by laser deposition. At the same time, P deposition technology will form a layer of plasma glow between layers, which hinders migration of electrons between layers also leads to a significant increase in electrical sistance.

Conclusions
The white LED was encapsulated by Cs4PbBr6 quantum dots and Gd2O3:Eu red ph phor as the light powder. Under the excitation of a GaN chip, the color coordinates of W-LED was (0.33, 0.34), and the color temperature was 5752 K, which was close to color coordinate and color temperature range of standard daylight. W-LED devices sh excellent current stability. With the increase in current and voltage, the luminescence tensity of quantum dots and phosphors increased by more than 10 times. However, It may be that the interlayer resistance was high because the spin-coated QDs were not annealed at high temperature. We also tried to use PLD deposition to form TiO 2 and CsPbBr 3 layers, and the J-V curve obtained is shown in Figure 8b. This was similar to the devices obtained by the one-step method, the PLD devices also had obvious optical response. The current density increased to 118.58%, but the current density was only 0.036 mA/cm 2 . The higher rate of current density increase may be related to the dense perovskite deposits. Very dense thin films can be formed by laser deposition. At the same time, PLD deposition technology will form a layer of plasma glow between layers, which hinders the migration of electrons between layers also leads to a significant increase in electrical resistance.

Conclusions
The white LED was encapsulated by Cs 4 PbBr 6 quantum dots and Gd 2 O 3: Eu red phosphor as the light powder. Under the excitation of a GaN chip, the color coordinates of the W-LED was (0.33, 0.34), and the color temperature was 5752 K, which was close to the color coordinate and color temperature range of standard daylight. W-LED devices show excellent current stability. With the increase in current and voltage, the luminescence intensity of quantum dots and phosphors increased by more than 10 times. However, the stability of the quantum dots was slightly less than that of phosphors when the current was on for a long time.
At the rotating speeds of 2000, 3000, and 4000 RPM, the thickness of the spin-coated TiO 2 layer decreased gradually. The film forming quality was the best at 3000 RPM, and the current density of the device was the highest at 3000 RPM. However, the increase in the current density increased with the increase in rotational speed. The thickness of CsPbBr 3 layer decreased, and the film quality increased with the increase in rotating speed. The best film quality, the highest current density, and the current increase range were obtained at 4000 RPM. The best device performance can be obtained by preparing a TiO 2 layer at 3000 RPM and a sCsPbBr 3 layer at 4000 RPM.
The photocatalytic device was constructed by means of toluene solution of spin coated quantum dots and PLD technology. The current density of the device was greatly improved when the light source was turned on. The device constructed by PLD technology improved by 118.58%, but its current density was much lower than that of the device constructed by solution spin coating method. This indicates that the perovskite layer prepared by PLD technology was denser and the interlayer resistance was larger.