Coplanar Capillary Plasma Electrode Discharge Flat-Panel Lamps Based on Porous Anodic Alumina

Abstract

A coplanar capillary plasma electrode discharge is a promising source owing to its superior performance. In this study, a coplanar capillary plasma electrode discharge flat-panel plasma lamp fabricated from porous anodic alumina and glass was designed to investigate its properties. Coplanar capillary plasma electrode discharge flat-panel lamps with porous anodic alumina dielectric layers were fabricated and investigated. Changing of the frequency and voltage of the power supply and altering of the thickness of the dielectric layers and the interval distance of the electrode were carried out to optimize the performance of the lamps by decreasing the spark-ignition voltage and enhancing the luminous efficacy. The luminance can exceed 7200 Cd/m² and the white light flux efficacy is more than 4.92 lm/W. The discharge process of capillary plasma electrode discharge was modeled and simulated using COMSOL Multiphysics. The electron density and temperature were also analyzed. The results show that small plasma jets were produced at the pores of the porous anodic alumina, which helped stabilize the plasma. The voltage in the gas gap changes sharply with the geometry of the porous anodic alumina dielectric layer, leading to a higher electric field. The spark-ignition voltage decreased. Small plasma jets increase the plasma uniformity. The electron density and electron temperature reach approximately 1.94 × 10⁸ m³ and 3.34 eV, respectively. A high electric field intensity produced at the porous anodic alumina validated the promising potential in related fields.

1. Introduction

The dielectric barrier discharge (DBD) has been widely used in several fields [1,2], such as polymer treatment [3,4], film preparation [5], lamps [6,7], and ozone synthesis. DBD produces a uniform discharge at atmospheric pressure and has the following configurations, as shown in Figure 1: volume barrier discharge (VBD), surface barrier discharge (SBD), and coplanar barrier discharge (CBD). The CBD devices can produce plasma on one side, which enhances the safety and flexibility of the treating space. Therefore, it is a promising structure for plasma medicine and surface treatment.

VBD has merits such as ease of establishment and a homogeneous discharge zone. It also has evident defects, such as a narrow discharge gap that restricts the treatment of large-volume substrates. In the design of flat-panel lamps with DBD, transparent and conductive materials (such as indium tin oxide (ITO)) are required to realize both conductivity and light release. To satisfy the requirements of mechanical pressure, the ITO glass requires sufficient thickness, which may neglect the optimization of the electric parameters. This restricts the enhancement of the luminous efficacy of VBD flat-panel lamps [7,8,9]. Concurrently, the higher price of ITO glass increases the price of the lamps. Therefore, it is necessary to establish a discharge mode that can solve these problems.

Using coplanar capillary plasma electrode discharge (CPED) flat-panel lamps, ITO glasses are not required, and the thickness of the dielectric layer can be adjusted as required [8]. Therefore, a flat-panel lamp based on the CBD structure was designed and fabricated in this study. The structures and discharge parameters were optimized to stabilize the discharge, decrease the spark-ignition voltage, and enhance the luminous efficacy.

The electron density in DBD is of the order of 10¹⁰ cm⁻³, which limits its use in several fields [1,2]. To solve this problem, Kunhardt proposed CPED [9,10], as shown in Figure 2. In CPED, several small, perforated holes are introduced. CPED devices introduce a variant of the barrier configuration to obtain plasmas with electron densities above 10¹² cm⁻³, which broadens plasma use [11,12]. The CPED has numerous capillary tubes, where the electric field is sufficiently high to produce numerous plasma jets and the electron density is enhanced by these jets. The instability of the glow-to-arc transition is restrained by the dielectric layer. A stable glow discharge with a high electron density was obtained at atmospheric pressure, which was also studied by Moskwinski [13]. In this study, the adaptation of CPED to aluminum surface cleaning was investigated. The CPED was divided into three capillary modes and a Monte Carlo simulation was used to investigate the energy distribution, nature of the plasma jets, and ionization process. Becker et al. [14] fabricated CPED devices using small-dimensional silicon or ceramics whose arrays offer good optical and electrical characteristics that are well-suited for applications in medical diagnostics, displays, and environmental sensing. Several structures of microplasma devices and their fundamental properties and applications were also discussed and studied. Koutsospyros et al. [15] applied CPED plasma for the destruction of hydrocarbons. The reactor volume, species residence time, specific energy input, and influent contaminant concentrations were also studied. The electron density reaches 10¹² cm⁻³, and the average electron energy is approximately 5–6 eV.

Moreover, the thick dielectric layer is a major disadvantage of CPED. Therefore, the identification of a thin dielectric layer with good dielectric properties is required. Metal oxide-based materials are widely utilized in different fields [16,17,18]. Porous anodic alumina (PAA) (Figure 3) has been prepared by anodization for several years [19,20,21,22,23,24]. A regular nanoscale porous polycrystalline structure appears on the PAA surface. Owing to their excellent dielectric material, PAA layers are suitable for CPED. DBD with PAA as a dielectric barrier at low pressures has been reported in previous studies.

Park [25] investigated a coplanar AC discharge between cylindrical electrodes with a nanoporous alumina dielectric. In their investigation, a modular dielectric barrier plasma device was designed, and an intense glow discharge was realized at 50 Torr neon with an alternating current waveform for a peak voltage of approximately 460–510 V. Hussain et al. [26,27,28] studied the characteristics of a radio frequency atmospheric-pressure DBD with dielectric electrodes. The discharge device was composed of PAA, which was obtained in a wet electrochemical process in 0.3 M oxalic acid at a constant voltage of 40 V. The current–voltage characteristics and millisecond images were used to distinguish the discharge modes.

COMSOL Multiphysics, a software tool for modeling and simulation in multiphysics, provides a user-friendly design environment. Simulations of DBD based on COMSOL have been performed in previous studies [29,30].

The structure of the coplanar CPED was similar to that of the CBD (Figure 1c), except that PAA acted as the dielectric layer. Similar work was performed in our previous study [8]. It is a promising plasma source to provide the space flexibility and enhanced properties. However, research on coplanar CPED with the PAA layers has not yet been conducted. To investigate the properties of coplanar CPED with the PAA layers, a coplanar CPED device with a PAA dielectric layer was fabricated and investigated in this study. A PAA layer was fabricated on an aluminum surface. The surface morphology was examined via scanning electron microscopy (SEM). Furthermore, a planar CPED flat-panel lamp with a PAA dielectric layer was investigated, and its luminous efficacy was examined. To further investigate the properties of the device, in this study, the discharge process of the coplanar CPED with the PAA layer was modeled and simulated using COMSOL Multiphysics 6.0. The main properties of the discharge, such as the electron density, electron temperature, and electrical field, were calculated and discussed.

2. Fabrication Process and Testing of Coplanar CPED Flat-Panel Lamps

The coplanar CPED flat-panel lamps are composed of CBD devices and glass. The fabrication process includes two steps: the fabrication of the CBD devices and the fabrication of the coplanar CPED flat-panel lamps.

2.1. Fabrication of CBD Devices

CBD devices were prepared in an anodizing apparatus, as shown in Figure 4. The apparatus consisted of a double cup with a cooler and a direct-current power supply (MS602D). Graphite and aluminum plates were connected to the negative and positive poles of the power supply, respectively, and both were immersed in the acid solution. Oxalic acid was used as the solution, which was cooled to 15 °C by the cooler. The power supply was set to a constant voltage of 40 V.

Figure 5 shows the coplanar CPED device composed of several sets of electrodes separated from each other and concealed in the dielectric layer. In this experiment, the electrodes and dielectric layer of the CBD devices were prepared by PAA with the width and spacing gap of the electrodes being 0.5 mm and 4.5 mm, respectively. The fabrication process of the coplanar CPED device was as follows:

(1)

A 70 μm pure aluminum foil (80 mm × 120 mm) was tightly bonded to a 3 mm thick ordinary glass under high temperature, high pressure, and an electric field. The mismatch between the expansion coefficients of the aluminum foil and ITO glass during the heat treatment process was solved in this manner.

(2)

Tar was coated on the aluminum foil using screen printing to obtain a mask film of the desired grid electrode structure. The tar-mask film was baked and solidified at 120 °C. The parameters in the tar coating process determined the interval distances and width of the CBD device. In this study, the interval distances of the CBD device were 0.5 and 1.0 mm. The width was taken as 1 mm.

(3)

The samples above were then electrochemically reacted in a 0.3 M oxalic acid solution at 15 °C until the reaction current was reduced from 300 mA to 0 mA, which indicated that all the aluminum not blocked by the tar-mask films was electrolyzed into Al₂O₃.

(4)

The tar-mask films were removed, and the samples were placed in an oxalic acid solution to continue the electrochemical reaction. After approximately 40 min, the thickness of the alumina layer reached the desired thickness, which was approximately 25 μm. The electrochemical reaction was complete, and a coplanar CPED device for CPED flat-panel lamps was obtained.

2.2. Fabrication of Coplanar CPED Flat-Panel Lamps

The structure and image of the coplanar CPED flat-panel lamp are shown in Figure 6. In the planar CPED flat-panel plasma lamp, an ordinary glass plate and the desired grid electrode structure (Figure 2) were used as the front and back plates. The interval distances of the electrodes were 0.5 and 1.0 mm. The width was 1.0 mm. The thickness of the dielectric layer was 25 μm. The luminous surface of the CBD flat-panel lamps was 70 mm × 110 mm.

To fabricate the coplanar CPED flat-panel lamps, the fluorescent powder was coated on the coplanar CPED devices, which were fabricated as described above, using the conventional screen-printing method. The lamps were packaged using the conventional packaging method.

(1)

White fluorescent powder was mixed with glue and coated onto the surface of the CBD devices via screen printing. When the mass concentration of the fluorescent powder was approximately 60%, the thickness of the fluorescent powder on the coplanar CPED device was approximately 65 μm.

(2)

A 3 mm transparent glass plate was used as the front plate, and the coplanar CPED device coated with fluorescent powder was used as the back plate. PbO was used as an adhesive to encapsulate the front and back plates in a box with a gap of 3 mm. Thus, the planar CPED flat-panel plasma lamp was fabricated.

2.3. Testing Process of CBD Flat-Panel Lamps

A planar CPED flat-panel lamp was connected to vacuum equipment and a gas supply system to obtain the base pressure and gas at different pressures. Pure Xe was selected as the discharge work gas at pressures of 20–120 Torr. Simultaneously, the two sets of electrodes shown in Figure 2 were connected to the two poles of the power supply. A DC square-wave pulse power supply was used in the experiment. The pulse frequency was 10 kHz, while the voltage was continuously adjusted in the range of 0–2500 V. The luminance of the lamp was measured using a luminance meter, and the current and voltage data were measured using an oscilloscope. The electric power of the lamps was calculated using the current–voltage method. The luminous efficacies of the lamps were also calculated. The influencing factors were analyzed, and the corresponding parameters were optimized.

2.4. Results and Discussion

The spark-ignition voltage was lower than the AC voltage when using a higher-frequency pulse power supply. A higher frequency and sharp variation in the electric field led to a lower spark-ignition voltage. In addition, a thinner dielectric layer can be discharged at lower breakdown voltages. The dielectric layer is the core component of the CBD devices, therefore, dielectric layers with different thicknesses have different capacitances, which affect the optimization of the CBD device circuit. In the process of preparing porous alumina by an electrochemical reaction, the thickness of the pore bottom depends on the voltage [16].

2.4.1. Influence of Interval Distance of Electrodes on the Spark-Ignition Voltage

The distance between the electrodes plays an important role in the DBD. When the interval distance is smaller, the electric field intensity will be stronger under the same discharge voltage. In addition, the free path of electrons is also reduced [9]. When a pulsed power supply is applied, the higher electromagnetic field generated by the rapid rise and fall of the pulse supply enhances the energy and free path of electrons, which improves the ion and ionization probabilities of the electron and gas molecules, respectively. Therefore, the interval distance must be optimized.

Figure 7 shows the pressures and spark-ignition voltages for devices with CBD discharge interval distances of 0.5 mm and 1.0 mm. Evidently, when the interval distance is 0.5 mm, the ignition voltage of the device is lower than that of the 1.0 mm distance. The free path and energy of the electrons are increased by the electromagnetic field generated by the pulsed power supply, which leads to the free path of electrons being significantly shorter than the electrode interval distance [8].

2.4.2. Influence of Pressure on Luminous Efficacy of Coplanar CPED Flat-Panel Lamps

Luminous efficacy is an important parameter of CBD lamps [8]. In this study, the lumen efficacy was determined using Equation (1):

$${\eta }_v=\frac{{\varphi }_v}{P},$$

(1)

where ${\eta }_v$ and ${\varphi }_v$ represent the luminous efficacy and flux, respectively, and P represents power consumption. Figure 8 shows the relationship between the luminous efficacy and voltage of the coplanar CPED flat-panel lamps under different pressures. Evidently, the luminous efficacy of the lamps increases with the voltage, and the maximum luminous efficacy is 4.92 lm/W. The luminous efficacy of the lamps with a gas pressure of 50 Torr was lower than that with a gas pressure of 20 Torr.

As the voltage increased, additional electric energy was converted into light energy, and the luminous efficacy of the lamps increased significantly. When the pressure was higher, the luminance of the lamps and luminous flux increased. However, at higher pressures, additional discharge channels in the lamps increased energy consumption, which significantly increased the electric power and decreased the luminous efficacy of the lamps.

3. Simulation of Coplanar CPED with Porous Anodic Alumina

3.1. Electron Density Distribution

To analyze the discharge characteristics of the PAA CPED, the discharge process was modeled and simulated using COMSOL Multiphysics 6.0. The simulation was performed with the 2D finite element method provided by the plasma module of COMSOL Multiphysics. The computational hardware had the following configuration: 11th Gen Intel(R) Core(TM) i9-11900 and 32 GB RAM. A physics-controlled mesh of normal size was built and applied. The reduced electron transport properties were applied in the plasma properties. The electron energy distribution function followed a Maxwellian distribution.

The structure of the coplanar CPED device with the PAA dielectric layer built in the simulation software is shown in Figure 9. The devices were composed of three parts: a porous anodic alumina layer, a gas gap, and a chamber wall (made of glass). The diameter of the pore and the thickness of the PAA bottom were 0.01 mm and 0.014 mm, respectively. The device parameters are listed in

Table 1. Parameters of the geometry of the porous anodic alumina in a CPED device.

Variables	dg/mm	dw1/mm	d/mm	d1/mm	Dp/mm	r
Description	Distance of gas gap	Width of electrode	Distance between electrodes	Thickness of PAA bottom	Diameter of pore	Ratio of depth to diameter
Values	0.50	0.01	0.02	0.014	0.02	1.5

. Anodes and cathodes were placed at intervals under the PAA layer. The power supply voltage and gas pressure were taken to 50 V and 50 Torr, respectively.

In the simulation, the dielectric constants of PAA and quartz were 9.1 and 3.75, respectively.

Ar was used as the discharge gas and the primary reactions are listed in

Table 2. Main reactions considered in porous anodic alumina CPED.

No.	Reactions
1	e + Ar => e + Ar
2	e + Ar => e + Ar*
3	e + Ar => e + Ar
4	e + Ar* => 2e + Ar+
5	e + Ar* => 2e + Ar+
6	Ar* + Ar* => e + Ar + Ar+
7	Ar* + Ar => Ar + Ar

, where Ar, Ar*, and Ar⁺ represent atoms, metastable atoms, and ions of Ar, respectively.

Furthermore, the main processes in the simulation were determined using Equations (1) and (2), which describe the transportation of the particles and the energy of the electrons. The relationship between the electric displacement vector and charges and the relationship between the electric field and potential are described in Equations (4) and (5).

$$\frac{\partial n_e}{\partial t}+\nabla \cdot {\stackrel{⃑}{\mathsf{\Gamma }}}_e=R_e-\left(\stackrel{⃑}{u}\cdot \nabla \right)n_e$$

(2)

$$\frac{\partial n_{\epsilon }}{\partial t}+\nabla \cdot {\stackrel{⃑}{\mathsf{\Gamma }}}_{\epsilon }+\stackrel{⃑}{E}\cdot {\stackrel{⃑}{\mathsf{\Gamma }}}_{\epsilon }=S_{en}-\left(\stackrel{⃑}{u}\cdot \nabla \right)n_{\epsilon }+\left(Q+Q_{gen}\right)/q$$

(3)

$$\nabla \cdot \stackrel{⃑}{D}={\rho }_q$$

(4)

$$\stackrel{⃑}{E}=-\nabla U$$

(5)

3.2. Results

To investigate the discharge process in the simulation, the evolution of the electron density, temperature, and electric field intensity of the CPED device with the PAA dielectric layer is shown in Figure 10, Figure 11 and Figure 12. The discharge process can be illustrated by the evolution. Electron density and temperature are the main parameters of coplanar CPED plasma. When a sinusoidal voltage is applied to the anode and cathode, the parameters change with the voltage. The voltage changes are determined using Equation (6).

$$V_{app}=V_m\cdot \mathit{sin}(\omega t),$$

(6)

where $V_{app}$, $V_m$, and $\omega$ are the applied voltage, peak voltage, and angular frequency, respectively. The $V_m$ was set at 300~500 V. The calculation of the sinusoidal voltage period was 1 × 10⁻⁴ s, which was one cycle. When a frequency of 10 kHz is applied, the time points selected occur at every 1/8 cycles (1/8 T) in one cycle, that is, (a) 1/8 T, (b) 1/4 T, (c) 3/8 T, (d) 1/2 T, (e) 5/8 T, (f) 3/4 T, (g) 7/8 T, and (h) 1 T.

The evolution of electron density in the coplanar CPED device is shown in Figure 10. In the first half of the cycle, the electron density near the anode significantly increased, whereas in the second half of the cycle, the electron density near the cathode increased. At 5.00 × 10⁵ s and 1.00 × 10⁴ s, the electrons were uniform, and the electron density was approximately 7.92 × 10⁷ m³. The highest electron density of 1.94 × 10⁸ m³ was obtained at 1/4 T.

The electron density was calculated to be 1/8 T and reached a maximum value at 1/4 T when the maximum voltage was applied. The same pattern was observed in the next process. The electron density was the indicator of the discharge occurrence.

The electron temperature evolution in the coplanar CPED device is shown in Figure 11. In the first half of the cycle, the electron density near the anode was significantly higher, whereas in the second half of the cycle, the electron temperature near the cathode increased. This result was consistent with the electron density. At 1/2 T and 1 T, the electron density distribution was uniform in the PAA zone and the electron density was 0.52 eV. The highest electron density of approximately 3.34 eV was obtained at 1/4 T.

The PAA with perforated holes functioned as the dielectric layer in CPED. The electron temperature in the PAA was much higher than that in the other part. Therefore, the PAA structure was shown to perform an important role in the CPED process.

To further investigate the role of the PAA in the CPED process, the electric field intensity was analyzed. The evolution of the electric field intensity in the coplanar CPED device is shown in Figure 12. In contrast to the results shown in Figure 11 and Figure 12, the electric field intensity between these zones was higher than that in the other zones. In the first cycle, the electric field intensity first increased. At 1/4 T, the electric field intensity was approximately 4.26 × 10⁷ m³, which was the highest value obtained. During 1/2 T and 1.0 T, the electric field intensity near the anode and cathode was higher at low values.

Evidently, the electric field intensity at the PAA part was much higher than in the other part, which is consistent with the results shown in Figure 11. The PAA dielectric layer provided a higher electric field intensity in the CPED process, which validates the promising potential in related fields, such as the plasma sources for material modification, plasma medicine, and many others.

4. Discussion

The voltage in the gas gap of the DBD devices (Figure 1a) was determined using Equation (7).

$$U_g=\frac{U{d}_g}{d+\frac{d_1}{{\epsilon }_1}+\frac{d_2}{{\epsilon }_2}},$$

(7)

where $d_1$, $d_2$, and $d_g$ are the thicknesses of dielectric 1, dielectric 2, and the gas gap, respectively, ${\epsilon }_1$ and ${\epsilon }_2$ are the dielectric constants of dielectric layer 1 and dielectric layer 2, respectively. The dielectric constant of the gas is assumed to be 1.00.

The capacitance of the dielectric layer is determined by Equation (8).

$$C=\frac{{\epsilon }_0{\epsilon }_{r}S}{d},$$

(8)

where ${\epsilon }_0$ and ${\epsilon }_r$ represent the absolute and relative permittivities of the dielectric layers, respectively.

In the PAA dielectric layer, the thickness of the pores varied, which led to a change in the voltage of the gas gap. The sharp change in the voltage in the gas gap causes a stronger electric field, as shown in Equation (9).

$$E_g=-\frac{\partial U_g}{\partial l},$$

(9)

where $E_g$ and $U_g$ are the electric field intensity and potential in the gas gap, respectively. The electric field intensity varied with the geometry of the PAA, which helped decrease spark ignition.

The results show that the PAA dielectric layer plays an important role in CPED devices. The electron density, electron temperature, and electric field intensity near the pores were higher than those in other parts. The electron distribution near the hole was uneven, similar to that of the plasma jets. The plasma jets facilitate the occurrence and stability of the coplanar CPED plasma.

To investigate the influence of PAA on the plasma in CPED plasma flat-panel lamps, the properties were calculated using the software. Figure 13 shows the plasma properties 0.01 mm above four pores of PAA in the CPED, namely, (a) the electron density and (b) the electron temperature.

As shown in Figure 13, the plasma concentration changed significantly above the pore zone. The electron density increased significantly, similar to that of the plasma jets. The plasma jets were small, which improved their uniformity. This uniformity assists in decreasing the electric power, which leads to an increase in luminous efficacy.

5. Conclusions

In this study, coplanar CPED flat-panel lamps with PAA dielectric layers were designed and fabricated. The ignition voltage and luminous efficacy of the lamps were measured and recorded. The luminous efficacies of the lamps reached 7200 Cd/m² and 4.92 lm/W. The discharge process of the coplanar CPED with a PAA dielectric layer was simulated using COMSOL Multiphysics 6.0. The electron density and electron temperature had similar distributions in one cycle. Furthermore, there was a fluctuation in the distribution of electron density and electron temperature, which behaves similarly to plasma jets. Plasma jets increase the uniformity of the plasma. A PAA dielectric layer is an effective material for decreasing the discharge voltage and stabilizing the plasma. The voltage in the gas gap changes sharply with the geometry of the PAA dielectric layer. A stronger electric field was produced between the pores of the PAA. In addition, the spark-ignition voltage decreased and the small plasma jets assist in stabilizing the plasma. The electron density and electron temperature reached approximately 1.94 × 10⁸ m³ and 3.34 eV, respectively. A high electric field intensity produced at the porous anodic alumina demonstrated the promising potential applications of CPED with a PAA in related fields, such as its application as an atmospheric pressure source in plasma medicine, surface modification, and lamps.