Linear plasmas, compared to small-area low-temperature plasma jets, offer a larger single-treatment area with a brush-like pattern, making them highly promising for various applications. This paper introduces the design of an adjustable low-temperature linear plasma source that operates under atmospheric pressure at 2.45 GHz. The design integrated microwave theory with dielectric barrier discharge principles, utilizing a resonant structure based on a microstrip power divider with one end open and the other shorted. The ground plane of the microstrip structure was replaced by a metal plate featuring a large groove. Argon gas was introduced from the short-circuited end and exited through the groove at the open end, where plasma excitation occurred. Experimental results demonstrated that the device, operating at atmospheric pressure, can achieve adjustable linear plasma widths ranging from 10 mm to 50 mm by varying the incident power between 30 and 100 W. Optical emission measurements confirmed the uniformity of the linear plasma, and the gas temperature 5 mm away from the discharge area remained at only 65 °C, even with a microwave incident power of 100 W. This study offers a novel approach to designing linear atmospheric pressure microwave plasma sources, with significant potential for diverse material treatment applications.
I. INTRODUCTION
Microwave atmospheric pressure plasma, characterized by high plasma density, high efficiency, significant ionization levels, and a greater abundance of active particles1–5 than plasmas sustained at lower frequencies, has applications in material synthesis,6 gas purification7 pollutant emission reduction,8–10 and hydrogen production.11–13 However, similar to other atmospheric pressure plasmas, microwave atmospheric plasma is limited by a short mean free path due to its high particle density. Additionally, the short microwave wavelength poses challenges in creating a large-area, uniformly intense electric field, thus restricting the overall plasma size.
Linear plasma arrays offer a significant advantage in enhancing discharge area and uniformity, effectively overcoming the limitations of small discharge regions and non-uniform discharges of typical atmospheric pressure microwave plasmas. This capability makes linear plasma arrays an attractive solution for industrial applications. Several studies have explored various designs to achieve this goal. For instance, Nowakowska et al.14 introduced transverse slots on a compressed waveguide, producing a linear atmospheric pressure plasma up to 100 mm in length, with gas temperatures ranging from 800 to 1250 K under incident powers of 300 to 850 W.
Chen et al.15 created a linear microstrip plasma array operating at 2.3 GHz, which was capable of sustaining a 110 mm long argon plasma at atmospheric pressure, though it required specialized gas intake equipment to maintain a pure argon environment. Kang et al.16 developed a linear low-power plasma source operating at 700 MHz based on a parallel plate resonator, generating a 20 mm long argon plasma. Eom et al.17 further contributed by generating a 30 mm argon plasma using a strip line structure at a resonant frequency of 2.41 GHz. However, many of these linear plasma sources encountered challenges such as limited plasma width, high operating temperatures, and reliance on noncommercial frequencies. These factors can restrict their applicability, increase operational costs, and add to system complexity.
Research has demonstrated that utilizing dielectric materials in the discharge region known as dielectric barrier discharges (DBD) can effectively generate large and uniform plasma discharges. This technique is widely adopted in the design of large-area uniform plasma sources at frequencies below 2.45 GHz.18–23 In contrast, reports on linear microwave-induced DBD plasma sources excited with frequencies at 2.45 GHz are relatively scarce. Kim et al.24 developed a low-power microwave atmospheric pressure plasma brush using a microstrip resonator structure and the DBD principle, generating a plasma brush with approximately 10 mm in length. Wu et al.25 designed a linear microwave plasma source utilizing parallel plate transmission lines and dielectric barrier discharge characteristics, achieving a 15 mm long plasma at 2.45 GHz. Unfortunately, these linear plasmas were too short and limit the application scenario.
In response to the aforementioned challenges, this article proposes a longer atmospheric pressure low temperature linear microwave plasma source, utilizing a microstrip transmission line resonator. This design significantly extended the length of the linear plasma and mitigated the formation of localized sparks or arcs by incorporating a power divider structure alongside the principles of dielectric barrier discharge. This approach enabled the generation of a large-area and low temperature plasma. Operating at a frequency of 2.45 GHz, the device allowed for the adjustment of the linear plasma length from 10 to 50 mm by varying the incident power from 30 to 100 W.
II. THEORETICAL BASIS OF A LINEAR MICROSTRIP TRANSMISSION LINE PLASMA SOURCE
As illustrated in Fig. 1(a), the plasma source consisted of two parts: a dielectric substrate coated with a copper layer and an aluminum ground. The dielectric substrate was composed of FR4 material ( , dielectric loss tangent = 0.02) with a thickness of 0.8 mm. Figure 1(b) was provided to show more clearly the assembling of this plasma source and its dimensions. In this structure, subx = 69.5, suby = 70, t1 = 11, t2 = 6.5, t3 = 7.3, t4 = 62.0, d1 = 3.0, d2 = 4.2, w = 70, l = 75.5, and h = 5.2 mm. An N-type connector was soldered to the aluminum ground, with its inner conductor passing through both the aluminum ground and the dielectric substrate. It was then connected to the copper layer at the feeding point, as shown in Fig. 1(a), using solder. Power was fed into the plasma source through the N-type connector via a 50Ω coaxial cable. It was taken into account that if the resonant cavity was excited through electric coupling, it would be difficult to ensure that the position of the inner conductor of the coaxial connector would be consistent with the simulation model when welding the coaxial connector to the feeding point, which would lead to a large variation between the actual results and simulations. Therefore, the excitation method we chose was magnetic coupling.
(a) Structure of the plasma source and (b) specific dimensions of the device.
The plasma source was based on a microstrip resonator, with one end open and the other short-circuited. The short end was achieved by using an aluminum screw with a diameter of 3 mm, where the metal screw's head was connected to the copper layer and the thread was embedded in the aluminum plate. The distance between the center of the screw and the feeding point was 5.3 mm. According to the microwave theory, when the microstrip transmission line terminal is open, the microwave is reflected at the open end, and the transmission line is in a standing wave state. At this time, the electric field strength at the open end is maximum, which helps to break down the gas and generate plasma. In order to obtain a higher power density distribution (electric field intensity) at the open end and reduce assembly difficulty, the discharge gap was suggested to be about 0.5 to 2 mm. Too large a discharge gap would reduce the electric field between the two plates, and the plasma discharge easily formed filaments.26 The gap g at the open end of the device was set at 1.2 mm.
Assuming the transmission line is lossless (or has low losses), the characteristic impedance is real. The output impedance lines and are chosen to provide the desired power division ratios.27 To achieve a uniform electric field distribution, the structure was designed as an equal power divider, setting meaning the widths of the two output transmission lines were equal. Discontinuities at the corners and cutoffs of the microstrip line introduce parasitic reactance, causing phase and amplitude errors in the electromagnetic wave signals and potentially affecting the input–output impedance matching. To minimize losses during microwave transmission and reduce the effects of discontinuities, the corners and cutoffs of the microstrip line were chamfered. Impedance was matched by controlling the position of the feeding point and adjusting the width and length of each copper strip segment.
(a) Structure of the printed circuit board and (b) equivalent circuit of the device.
(a) Structure of the printed circuit board and (b) equivalent circuit of the device.
To form a resonator that facilitates achieving the maximum electric field at the open end for plasma excitation, thereby breaking down the gas and generating plasma, we let and , where is the wavelength of the electromagnetic wave propagating in the medium at an incident frequency of 2.45 GHz. It should be noted that the above formulas are mainly derived from theoretical derivations and approximate calculations.
A corresponding 3D model was established for simulation based on the dimensions above. The electrical conductivities of Cu and Al used in the model were 5.8 107 and 3.8 107 S/m, respectively, and the dielectric constant of the FR4 material used was 4.4. The electric field obtained at the open end was not uniform and did not reach the minimum ionization threshold28 required to break down argon. Therefore, the device dimensions need to be further optimized and adjusted using the finite element method to ensure a uniformly distributed and sufficiently strong field strength at the open end. The final optimized parameters are as follows: l1 = 5.3, l2 = 17.4, l3 = 36.1, l4 = 8.7, w1 = 9.1, w2 = 13.5, w3 = 13.1, w4 = 50.1, w5 = 2, c = 9.9, and c1 = 2.0 mm.
Figure 4, shows the electric field distribution at the discharge gap of the open end in the optimized structure before the plasma was excited when the microwave frequency was 2.45 GHz and the incident power was 1 W. It can be observed that an approximately uniform electric field distribution can be formed within an area of 2 with a maximum electric field strength reaching 1.96 104 V/m. Kang et al.16 successfully excited a uniform 20 mm long linear argon plasma at atmospheric pressure. According to their simulation results, an incident power of 1 W produced a maximum electric field strength of 5.5 × 102 V/m, demonstrating that the electric field generated by our device should be sufficiently strong and has the potential to excite and sustain plasma at atmospheric pressure. It also showed that the ratio of the maximum to minimum electric field intensity at the open end was approximately 1.1, indicating that the applied electric field is almost uniform16 before plasma excitation.
Electric field distribution of the discharge gap at the open end before plasma excited.
Electric field distribution of the discharge gap at the open end before plasma excited.
III. EXPERIMENTAL SETUP
In order to investigate if this device could generate linear plasma under atmospheric pressure, a microwave excitation system as shown in Fig. 5 was built. A microwave solid-state source (WSPS-2450-1K-CCWA, Wattsine Electronic Technology Ltd.) was used to supply power. A circulator ((L00PE22DC40A10N, Euler Microwave Element Ltd) with a matched load was connected next to the microwave source to absorb the energy reflected by the system and protect the microwave source. A microwave power meter (AV2433, the 41st Institute of China Electronic Technology Group Corporation,) was connected with a double directional coupler (MPHWG22BDCA01, Mapping Huineng Technology Ltd.) to measure the incident and reflected power of the system. A spectrometer (AvaSpec-ULS2048-5-EVO, Avantes Technology Ltd.) was employed to obtain the emission spectra of the generated plasma discharge. The optical fiber probe was placed 15 mm horizontally away from the plasma. The working gas used in the experiment was argon with purity greater than or equal to 99.99%, and its flow rate was measured by a flowmeter (MF5706, Siargo Ltd).
IV. EXPERIMENTAL RESULTS
A vector network analyzer (Agilent N5230A) was used to assess the port matching of the device. A coaxial cable connected the analyzer to the plasma source, and |S11| was measured in the single-port mode. Figure 6 presents a comparison between the simulated and measured reflection coefficients |S11|. At the resonant frequency of 2.45 GHz, both values were approximately −30 dB, indicating that theoretically, before plasma generation, about 99% of the input energy could be transmitted to the load with minimal energy loss.
Figure 7 illustrates the plasma morphologies observed at 2.45 GHz with a gas flow rate of 3 l/min under different incident power levels. All images presented in this study were captured using a camera (DSC-RX10M3, Nikon) positioned 10 cm from the plasma source at the same height (as shown in the Fig. 5), with an exposure time of 1/1000 s. The experiments demonstrated that a linear plasma with a length of 10 mm could be initiated and sustained at an incident power of 30 W. Increasing the incident power to 50 W extended the plasma length to approximately 15 mm, and a significant increase to 30 mm when the incident power is 70 W. Notably, when the incident power reached 100 W, the plasma discharge covered the entire gap between the microstrip and the ground plate, with the plasma length extending to 50 mm. Further increases in incident power resulted in negligible changes in the plasma morphology.
Photographs of linear plasma morphologies at different powers (gas fiow:3 l/min).
Photographs of linear plasma morphologies at different powers (gas fiow:3 l/min).
The plasma width expands with increasing power, likely due to a stronger electric field at the open end as incident power rises, enhancing the ionization efficiency of gas molecules. This results in an increase in electron density, indicating that more gas molecules are ionized, thereby extending the plasma length.
The emission spectra of the plasma discharge at different incident power levels were measured and are presented in Fig. 8. To enhance clarity, the spectra were measured and displayed in segments, with an acquisition time of 100 ms for the wavelength range of 280–420 nm, and 2 ms for the range of 720–860 nm. As shown in Fig. 8, the spectral lines corresponding to argon species (720–860 nm) dominate the spectrum. Additionally, spectral lines from species, such as and , were also observed.
Emission spectrum of the plasma source at different powers (gas fiow:3 l/min) (a) 280–420 nm and (b)720–860 nm.
Emission spectrum of the plasma source at different powers (gas fiow:3 l/min) (a) 280–420 nm and (b)720–860 nm.
The plasma emission spectra show that higher incident power intensifies the spectral lines of active particles, suggesting that more active particles are generated. This is likely due to increased collisions between energetic particles in the argon plasma and oxygen molecules and water vapor in the surrounding air. The presence of more active particles, such as OH and O, enhances the plasma's effectiveness in various applications, as these active species are critical for processes like sterilization29,30 and pollutant treatment.31
The uniformity of the plasma discharge along the length of the gas channel was assessed using emission spectra. The midpoint of the linear plasma, 25 mm from the edge of the copper layer at the open end, was established as the central reference. Emission spectra were then measured at 10 mm intervals in both directions from this center. The relative intensity variations of the spectral lines for (750.4 nm) and radicals (309 nm) at different positions are depicted in Fig. 9. At an incident power of 100 W, the maximum variations in relative intensities for OH and Ar were 329.93 and 456.69, respectively. At 70 W, these maximum variations were 136.76 for OH and 309.77 for Ar.
Ar (750.4 nm) and OH free radical (309 nm) intensity along the discharge gap (microwave frequency 2.45 GHz, flow3 l/min, and incident power 70 and 100 W, respectively).
Ar (750.4 nm) and OH free radical (309 nm) intensity along the discharge gap (microwave frequency 2.45 GHz, flow3 l/min, and incident power 70 and 100 W, respectively).
The effects of varying incident power on gas temperature, plasma ambient temperature and electron density were also investigated. The gas temperature of the argon plasma was determined from the emission spectrum of , using LIFBASE software for calculations, referring to the overall thermal state of the neutral gas in the plasma.32 Additionally, as shown in Fig. 10, the ambient temperature was measured using a fiber optic thermometer (FOS-TG-01-P-0-000, Omega Engineering).
Schematic of ambient temperature measurement using a fiber optic thermometer.
Figure 11 presents the gas temperature and temperature at a spot 5 mm horizontally away from the discharge center (referred to as the ambient temperature in the following text) after plasma excitation, at different incident power levels. It was observed that both the gas temperature and the ambient temperature increased with rising incident power. Specifically, at an incident power of 100 W and a gas flow rate of 3 l/min, the gas temperature at the discharge region reached approximately 750 K. At this power level, the ambient temperature, measured using a fiber optic thermometer (as shown in Fig. 11), was approximately 338 K (65 °C). This indicated indirectly that it has promising potential in various fields, including some temperature sensitive fields.
Gas temperature and ambient temperature at different incident power levels.
Table I presents the electron density of plasma discharges under various incident power conditions. The data indicated once again that increasing incident power could lead to higher electron density. Therefore, the microwave incident power could effectively change the discharge length and its properties. One could adjust the incident power to regulate or obtain the plasma discharges with properties that they want just by adjusting the microwave incident power.
Electron density (cm−3) at different incident power (gas flow: 3 l/min and microwave frequency: 2.45 GHz).
Incident power (W) . | Electron density (cm−3) . |
---|---|
30 | 3.45 × 1015 ± 3.63 × 1014 |
50 | 1.87 × 1016 ± 3.22 × 1015 |
70 | 1.93 × 1016 ± 4.66 × 1015 |
100 | 2.37 × 1016 ± 6.07 × 1015 |
Incident power (W) . | Electron density (cm−3) . |
---|---|
30 | 3.45 × 1015 ± 3.63 × 1014 |
50 | 1.87 × 1016 ± 3.22 × 1015 |
70 | 1.93 × 1016 ± 4.66 × 1015 |
100 | 2.37 × 1016 ± 6.07 × 1015 |
V. CONCLUSION
This paper presented the design of a novel adjustable low-temperature linear microwave plasma source operating at atmospheric pressure and 2.45 GHz, based on a microstrip resonator structure and dielectric barrier discharge. Experiments showed that with this device, an incident power of 30 W could trigger and stably maintain a 10 mm plasma discharge, while an incident power of 100 W can achieve a 50 mm linear plasma. Measurement of spectra at different positions showed that the generated linear plasma was uniform. Adding microwave incident power not only extended the length of plasma, but also increased the electron density, gas temperature, and active species concentrations. The electron density was up to ×1016 cm3 and the gas temperature was 750 K with 100 W of incident power, while the ambient temperature 5 mm away from the plasma discharge area was only 65 °C. This linear plasma source is simple, has adjustable properties and morphology, and is low-temperature, which renders this device highly versatile for processing various materials across diverse domains, thus facilitating a broad spectrum of potential applications.
ACKNOWLEDGMENTS
This research was financially supported by the Natural Science Foundation of Sichuan (2023NSFSC0707) and the Institutional Research Fund from Sichuan University (2023SCUH0003).
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Ao Qu: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Methodology (equal); Software (equal); Writing – original draft (equal). Yan Chen: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal). Nian Zhang: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Resources (equal). Li Wu: Conceptualization (equal); Funding acquisition (equal); Methodology (equal); Resources (equal); Software (equal); Supervision (equal); Writing – review & editing (equal).
DATA AVAILABILITY
The data that support the findings of this study are available within the article and its supplementary material.