1. Introduction

Perovskite quantum dots (PQDs) are semiconductor nanocrystalline materials that have drawn significant attention in recent years [1] due to exceptional properties such as high quantum yield, adjustable band gap, wide color gamut, and high purity. As a result, they have found applications in the fields of ultra-clear display technology and high-end lighting. Due to the rapid development of perovskite material in display devices, patterning techniques have been extensively researched to optimize the microstructure machining of perovskite materials. In the past few years, several patterning methods have been developed for quantum dots (QDs) such as inkjet printing, transfer printing, masking technique, nanoimprinting, and lithographic patterning [2–7]. Although these methods are efficient and offer a high degree of patterning precision, they rely on stencils and high-precision hardware, which has limited their large-scale adoption. Recently, a faster and more convenient laser direct writing (LDW) perovskite patterning technology has emerged, which is maskless and programmable [8]. However, the technique has a propensity to cause undesirable effects such as phase transition and damage to the perovskite [9–11]. Therefore, there is a need for developing new techniques for patterning PQDs that are scalable and precise.

Liquid crystal displays (LCD) are rarely used for transparent displays because their transmittance is only 20%, which is caused due to optical elements such as by polarizers and color filters. Although LED technology is used in transparent displays, it has a transmittance of 40% [12], which does not meet the requirements in many applications. Notably, indium tin oxide (ITO) coated glass has a transparency of more than 70% in the visible region, which is an ideal substrate for transparent displays [13,14]. Patterns can be created on the glass substrate by laser etching and then filling QDs subsequently. This approach allows the QDs to be encapsulated and preserved without increasing device thickness, effectively improving stability and lifetime while achieving a transparent display. However, the direct laser processing of quartz glass surface is relatively difficult. Quartz glass exhibits low absorption to ordinary pulsed lasers and the absorption is generally around 10% [15]. At present, the mainstream etching methods of quartz glass are mainly divided into laser-induced front side etching (LIFE) and laser-induced back side dry etching (LIBDE). In the case of LIDBE, etching is performed by adding an auxiliary layer to the surface of quartz glass [16–18]. Another method is called laser-induced plasma etching, where laser action is carried out on the contact surface of quartz glass and target material, through laser excitation metal target plasma to etch quartz glass. It is a common technique used to process transparent, brittle, and hard materials. Owing to the low etching threshold, this method causes less damage and has a relatively smaller number of cracks, raw edges, and debris compared to other methods, which is suitable for the precision machining and micro-machining of transparent materials [19–22].

In this paper, patterning of PQDs is demonstrated based on the laser-induced plasma etching approach. Experiments with different parameters are designed to arrive the optimal processing parameters in order to improve the accuracy of patterning. For the metal target, three metals are used, namely, zinc [23], molybdenum [24], and titanium [25] with widely different physical properties to compare the etching effect. In order to cover the substrate with a constraint layer to prevent undesired QDs from appearing outside the pattern, a polyvinylidene fluoride (PVDF) film is used as the constraint layer. PVDF is chosen because it has good impact resistance, corrosion resistance and weathering resistance, which results in the ablated holes being uniform with precise dimensions. Finally, a PQDs dot matrix is fabricated with patterns features of 10-20 µm diameter and 100 µm pitch, which mimic a micro-display. In addition, with the submicron level patterning processing on transparent materials and the fluorescence property of PQDs under UV-light, this patterning method can be effectively used for anti-counterfeiting or encryption applications.

2. Material and methods

The experiment is accomplished in six steps: a) synthesis of perovskite quantum dots; b) synthesis of constraint layer polymers; c) fabrication of samples covered with polymer films; d) laser-induced plasma etching; e) quantum dot filling; f) polymer removal to complete patterning.

2.1 Synthesis of perovskite quantum dots and PVDF films

The thermal injection method was used to synthesize the three all-inorganic perovskite quantum dots: CsPbBr₃, CsPbBrI₂ and CsPbClBr₂ [26]. The concentration of quantum dots is 0.135 mmol/ml. The solvent is toluene and the ligands are oleic acid and oleylamine. Next, PVDF solution was prepared. The 1.26 g PVDF powder (procured from Macklin) powder is dissolved in 10 ml dimethylformamide (DMF, procured from Aladdin) solvent and reacted in a magnetic heating agitator at 50 °C, 700 rpm for 10 hours. The solution was prepared and poured evenly on to a slide with dimensions of 2 × 2 cm. It was then cleaned with acetone and placed on a scraping table. The parameters of thickness and speed were set for scraping, as shown in Fig. 1(a) Subsequently, the slides were placed in a drying oven at a constant temperature of 50 °C to form a thin film, and finally the substrate covered with PVDF film was obtained. The thickness of the PVDF film used in the experiment is 0.3 mm. The metal targets: zinc (purity > 99.995%, length =35 mm, width = 35 mm, and thickness = 2 mm), molybdenum (purity > 99.99%, length =35 mm, width = 35 mm, and thickness = 2 mm), and titanium (purity > 99.99%, length =35 mm, width = 35 mm, and thickness = 2 mm, all procured from Zhongnuo Advanced Material Technology Co., Ltd).

 

Fig. 1. (a) Diagram of PVDF film fabrication (b)Principle of the experimental setup. The order of sample placement from top to bottom is glass substrate, PVDF film and metal target.

Download Full Size | PDF

2.2 Laser-induced plasma etching and patterning process

Figure 1(b) shows the experimental setup for the laser-induced etching procedure. The pulsed laser has an operating wavelength of 532 nm, a pulse width of < 1.8 ns and a beam diameter of 300 ± 100 µm. The laser is introduced into the scanning galvanometer system through an attenuator and reflector. The laser beam is output by the galvanometer after the optical path adjustment. Herein, the LIBDE etching method is employed, where laser is focused on the upper surface of the metal. The sample is placed in the center of the moving stage from top to bottom in order of SiO₂ substrate, constraint layer (PVDF film) and metal target. The etching pattern is designed in the galvanometer software system. When the system is powered on, the laser penetrates the glass substrate and PVDF film, and directly radiates on the metal target. The generated plasma instantly burns the middle layer film and the glass substrate, thereby achieving the etching effect. After etching, patterned PVDF films are obtained. The prepared QDs solution is then spin-coated onto the substrate, with the rotating speed and coating time being 2000rpm and 60 s, respectively. The sample is then placed in a closed drying box for 24 hours until the quantum dot solution is completely dry. Finally, the PVDF film is peeled off.

3. Results and discussion

In order to obtain a finer pattern, each parameter of the etching section needs to be adjusted and optimized. The etching process in the experiment involves a photoelectric mechanism, which is based on the inverse bremsstrahlung absorption [27]. When the laser power density reaches a certain threshold, the action on the metal target produces plasma. The microscopic mechanism can be divided into two steps. In the first step, the electrons on the metal surface absorb photon energy through the inverse bremsstrahlung absorption. The absorbed electrons then transfer energy to the metal lattice through electron-phonon interaction, and the lattice is destroyed to generate Joule heat [28]. In the second step, further ionization and heating of the sputtered material occurs when ions with a certain energy collide with molecules and atoms, which causes the photoionization of atoms and molecules to exist in an excited state. Simultaneously, multiphoton ionization of atoms and molecules in the ground state also appear. The aforementioned mechanisms cause the generation of plasma on the surface, which continues to absorb the remaining energy of the laser and rapidly expands outward to engender a plasma shock wave. These events result in the bombardment of the polymer surface, penetration, and further ablation of the quartz glass. When the pulsed laser energy is sufficient, the resulting plasma can instantly burn the film due to its high temperature [29,30], which in turn causes the substrate surface to evaporate and melt [31]. Based on this, a model can be established, where the process is divided into two steps: a) production of heat by the direct laser beam irradiation on the target material, and b) thermal ablation of the polymer film and etching of the quartz substrate. The energy distribution of the laser can be calculated from the following equations:

 

Fig. 2. (a) Simulation results of the etching process heat flow of three metal targets under the same laser conditions. Inset shows the etching position. (b) Ablation depth of PVDF film. Inset shows the animation screenshot of the ablation depth.

Download Full Size | PDF

The second part of the model involves the simulated heat ablation of PVDF film. Based on the results derived from Eqs. (1) and (2) the heat source. The using PVDF parameters is listed in Table 1. With these results and parameters, a simulation can be established of the ablation depth. From Fig. 2(b) it can be seen that the ablation depth varies linearly in the range of first few seconds, and the ablation depth reaches 0. 14 cm after 14 seconds. Therefore, the desired film thickness can be determined by the heat flow generated by the laser energy and the time of action.

Table 1. Polyvinylidene fluoride (PVDF) material parameters

View Table

The plasma excited by three metal targets with a mono-pulse laser were experimentally observed using a high-speed camera and the results were compared, as shown in Fig. 3. The threshold laser energy of 90 µJ was obtained from several experiments. In the right half of Fig. 3(a), the right half of Fig. 3(b), and the right half of Fig. 3(c), it can be seen that the ionization and evaporation occur when the laser captured by the high-speed camera directly reaches the surface of three metal targets. The upper and lower figures represent the moment the pulsed laser reaches the metal surface and the action process. It is also found that the plasma excited by laser irradiation on zinc is stronger than the other two targets. In order to visualize the etching effect, the galvo scanner is set to a marking speed of 0.1 mm/s and an output frequency of 80 Hz. The ablation pits of the three targets after 10 consecutive etches are characterized by confocal laser scanning microscopy (CLSM), where it can be seen that the etching effect of target material zinc (Fig. 3(a)) is stronger than that of target molybdenum (Fig. 3(b)) and titanium (Fig. 3(c)). Additionally, morphology is smoother and more uniform in the case of the zinc target. The average etching depths and ultimate depths of the zinc, molybdenum and titanium targets are 0.64 µm, and 3.97 µm, 0.1 µm and 1.25 µm, 0.49 µm, and 1.51 µm, respectively, as can be seen in Fig. 3(d, e, f). The ultimate depths can be obtained from the lowest point in the depth representation, while average etching depths are determined from the average height difference between the unetched area and the etched area in the CLSM test data. After subsequent filling of the QDs, it is observed that the pattern made with zinc as the target material is better in terms of uniformity of appearance and size. As a result, zinc is chosen as the target material. Next, three sets of etching cycles are chosen (10, 20 and 30 times) and the CLSM images are captured for each setting, as shown in Fig. 3(a, g, h). It can be seen that the lines are more uniform and continuous as the number of etchings increases. From Fig. 3(d, i, j), it can be seen that the average depth of the dent increases slowly according to the numerical difference between the blue line and the red line, which is 0.64 µm,1.29 µm, and 1.79 µm respectively. However, the ultimate depth increased rapidly, from 3.97 µm to 13.56 µm. The aperture diameter is basically maintained at 150 µm. This indicates that increasing the number of etchings does not lead the pits to lateral spreading, but only increases the vertical depth. Experiments also reveal that repeated etching does not affect the accuracy of the pattern. Hence, the number of repetitions can be chosen depending on the desired surface depth.

 

Fig. 3. Characterization of etched morphology with different metal targets and different etching frequencies. CLSM images of various targets with 10 etching cycles and the moment of plasma formation captured by a high-speed camera (a) Zn (b) Mo and (c) Ti. (d) Cross-sectional etching depth of targets after 10-fold etching (d) Zn (e) Mo and (f) Ti. Etching characterization of zinc with (g) 20 etching cycles and (h) 30 etching cycles. Cross-sectional etching depth of Zn target with (i) 20 etching cycles and (j) 30 etching cycles.

Download Full Size | PDF

In order to explore the feasibility of this method, a dot matrix series pattern is fabricated, and computer positioning is utilized to achieve three colors on the same plane. Figure 4(a) shows the steps to make a QD array. The spacing is 100 µm, and the pore size is about 10-20 µm. First, a high-speed camera records the moment of pulsed laser action on the metal target penetrating the film. The green boxes in the picture indicate the moment of etching in progress, and the white boxes indicate the holes formed in the PVDF film after the etching. Next, the QDs are spin-coated and filled into the etching holes on the substrate. The PVDF film is finally peeled off. A dot array consisting of PQDs can be observed under a microscope, as shown in Fig. 4(c). The fluorescence effect under UV-lamp excitation can be seen in Fig. 4(d). For the purpose of achieving multi-color functionality in the same plane, the computer translation table is combined with microscope positioning, as shown in Fig. 4(b). This process is repeated three times to obtain the tricolor display.

 

Fig. 4. (a) Fabrication process of PQDs dot matrix. Step.1, dropping the PQDs onto the patterned sample surface; step.2, spin-coating; step.3, peel off the PVDF film (b) Multiple positioning to achieve tricolor. Filling with PQDs under (c) halogen lamp and (d) UV-lamp.

Download Full Size | PDF

The accuracy of patterning needs to consider the target material, film thickness, laser energy, and etching frequency The stress field generated by the plasma impacting the film and the glass slide can cause cracks in the original film that is tightly attached to the substrate [33], resulting in blurred patterns. The laser energy is systematically increased from 10 µJ while etching the substrate using a 0.3 mm PVDF film to ascertain the minimal energy value. Through comparative experiments, the minimum value denotes the energy at which precise penetration of the PVDF film results in the formation of a pattern on the substrate. This energy is determined to be the minimum energy value. To make the stress field as small as possible, the minimum energy is set as a threshold while satisfying the pattern formation condition. The minimum energy value in the current experiment turns out to be 90 µJ. When it comes to the selection of the target material, zinc yields the best results, where regular and smooth patterns are observed on the film. Finally, three colors of PQDs patterns are generated on the same plane based on the laser-induced plasma etching. Figure 5(a) shows some letters and drawings under the UV-light. Interestingly, in addition to the application of this patterning method in display devices, it can also be used in the field of transparent display and anti-counterfeiting. Under an appropriate selection of materials and parameters, the vertical depth of the etching can be controlled below 1 µm. It is challenging to visually distinguish this subtle height difference on transparent materials, making this suitable for security markings that use submicron anti-counterfeit engraving. Figure 5(b) depicts the shooting effect of the same sample produced under different lighting conditions. The yellow arrow below the photo indicates that the light intensity goes from weak to strong. Through the transparent glass, the school badge can be seen on the back of the sample, but the “NUST” pattern printed on the glass cannot be observed. Since PQDs are fluorescent substances, the luminescent patterns can be clearly seen under the anti-counterfeiting light. A low-intensity normal light source is also introduced in the right image of Fig. 5(b) to observe the transparency effect.

 

Fig. 5. (a) Fluorescent images of letters and patterns. (b) Transparency effect of samples under different lighting conditions, the same picture on the right is under the anti-counterfeit light. A normal light source is used to show the transparency effect. (c) PL of the three original quantum dot solutions. (d) PL after 15 days of patterning. (e) Transmittance of patterned RGB PQDs. (f) Fluorescence lifetime of patterned RGB PQDs (after double exponential fitting).

Download Full Size | PDF

Photoluminescence (PL) tests were performed on PQDs before and after patterning, and no reduction was observed in the photoluminescence efficiency of this patterning method. The luminescence spectra of original CsPbBr₃, CsPbBrI₂, and CsPbClBr₂ PQDs are shown in Fig. 5(c), comparing with the PL after fifteen days of patterning in Fig. 5(d). It can be seen that the luminescence peak and intensity of the quantum dots did not change significantly after filling. This is attributed to the fact that the proposed method avoids laser interaction with PQDs and protects the PQDs under the substrate surface. Additionally, the light transmittance and the fluorescence lifetime of the samples is examined after patterning. The transmittance of the three colors after patterning is shown in Fig. 5(e), with daily UV resistant glass as a reference. It can be seen that the light transmittance of all three RGB colors in the visible light range is above 60%, meeting the technical requirement of 40% transmittance for transparent displays. In Fig. 5(f), the fluorescence lifetime was obtained by a time-correlated single-photo counting system, and a double exponential fitting was performed to calculate the fluorescence lifetime. The fluorescence lifetimes of R, G, and B colors were found to be 26.17 ns, 9.26 ns and 5.14 ns, respectively. All inorganic lead halide perovskites have high luminous efficiency, but they contain heavy metals that are harmful to the environment. The proposed method not only satisfies the requirements for transparent displays, but also provides a new approach for environmental protection.

4. Conclusion

The proposed laser-induced plasma etching method can encapsulate PQDs below the substrate surface at submicron depths. In contrast to other patterning techniques, this encapsulation strategy shields the PQDs from the environment and confines them within the etched glass. Additionally, this approach enables other use cases for PQDs such as transparent displays and anti-counterfeiting applications. Herein, experiments were designed with different parameters to improve the accuracy of the patterns to minimize burrs, debris and other problems caused by stress fields. A fluoropolymer with good physical and chemical properties was finally selected as it could produce smoother and finer patterns. Among the various metal targets, zinc was selected as it produced a more intense plasma and performed better in terms of uniformity and controllability of the pattern. The minimum diameter of 10 µm dot matrix and a 3 × 3 mm size pattern was realized in this work, with the transmittance being above 60%. The depth of the etched patterns was adjusted by the number of etching cycles, which increased the vertical ultimate depth, allowing more quantum dots to be filled in order to increase the luminance without causing a degradation of the pattern accuracy. This novel approach for patterning PQDs shows great potential for applications in the field of transparent displays, while also opening up avenues in related research fields.