Gas breakdown in radio-frequency field within MHz range: a review of the state of the art

Wei JIANG; Hao WU; Zhijiang WANG; Lin YI; Ya ZHANG

doi:10.1088/2058-6272/aca648

Plasma Science and Technology > 2022 > 24(12): 124018. > DOI: 10.1088/2058-6272/aca648

Wei JIANG, Hao WU, Zhijiang WANG, Lin YI, Ya ZHANG. Gas breakdown in radio-frequency field within MHz range: a review of the state of the art[J]. Plasma Science and Technology, 2022, 24(12): 124018. DOI: 10.1088/2058-6272/aca648

Citation:

PDF (1656 KB)

Gas breakdown in radio-frequency field within MHz range: a review of the state of the art

1.
School of Physics, Huazhong University of Science and Technology, Wuhan 430074, People's Republic of China
2.
International Joint Research Laboratory of Magnetic Confinement Fusion and Plasma Physics, State Key Laboratory of Advanced Electromagnetic Engineering and Technology, School of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, People's Republic of China
3.
School of Electronics and Information Engineering, Hubei University of Science and Technology, Xianning 437100, People's Republic of China
4.
Department of Physics, Wuhan University of Technology, Wuhan 430070, People's Republic of China

More Information

Corresponding author:
Lin YI, E-mail: yilin@hust.edu.cn
Received Date: September 21, 2022
Revised Date: November 21, 2022
Accepted Date: November 24, 2022
Available Online: December 05, 2023
Published Date: December 22, 2022

Graphical Abstract

Abstract

Abstract

Low-temperature plasmas (LTPs) driven by 1–100 MHz radio-frequency (MRF) are essential for many industrial applications, and their breakdown characteristics are different to that of direct current (DC) breakdown. This review seeks to understand the state of the art of electric breakdown in the MRF field and provide references for related basic and applied research. We have given a brief history of research into MRF-driven breakdown, including Paschen curves, the corresponding discharge modes and parameter spaces, and the evolution of the parameters during the breakdown process. It is shown that the focus has been transferred from the breakdown voltage and V-I characteristics to the evolution of plasma parameters during the breakdown, both in experiments and simulations. It is shown that many fundamental and applied problems still need to be investigated, especially with the new global model and the incorporation of the external circuit model.
- gas breakdown,
- RF plasma,
- Paschen curve

FullText(HTML)

1. Introduction

When a hypersonic vehicle is flying in the atmosphere, it interacts with the atmosphere and forms a bow-shaped shock wave on the head of the vehicle. The air temperature rises sharply, causing the atmosphere to dissociate and ionize, and a plasma sheath is generated around the aircraft, which affects the communication between the aircraft and the ground measurement and control system. This phenomenon is called a black barrier [1–5]. In order to fully understand the interaction mechanism between plasma and electromagnetic waves and solve the black barrier problem, inductively coupled plasma (ICP) is generated by a ground simulation device, and the plasma parameters are diagnosed. The main parameters of plasma that affect the transmission of electromagnetic waves are its electron density and collision frequency [6], so the diagnosis of these two parameters is of great significance. Microwave diagnosis [7–11], laser interference diagnosis [12], probe diagnosis [13–15], and spectrum diagnosis are all commonly used plasma diagnosis methods [16].

Langmuir probe diagnosis is the most basic diagnosis method for plasma parameters [17]. The probe with applied voltage is situated deep in the plasma, and the relationship between the current and the voltage of the probe is measured, and the volt-ampere (V-A) characteristic curve of the probe can be obtained [18]. According to the special points on the curve, the plasma density, electron temperature, plasma potential and floating potential can be calculated. In addition, H Kokura et al studied the plasma absorption probe (PAP) [19], which is based on the characteristics of resonant absorption of surface waves excited by the probe head. Even when the probe is affected by plasma, it can still be used for the diagnosis of plasma electron density. Since the probe needs to go deep into the plasma, when the plasma temperature is high, the probe will be burnt out. Therefore, in the research of high-temperature plasma, probe diagnosis is not appropriate, and noncontact diagnosis methods are generally used.

Laser interference diagnosis is a commonly used noncontact diagnosis method. This method regards plasma as a medium. When the laser is transmitted in the plasma, the refractive index of the plasma is related to its electron density. Therefore, the electron density can be obtained by measuring the refractive index of the plasma [20]. Due to the high frequency of lasers, laser interferometric diagnosis is generally used to diagnose plasma with high electron density. Laser interference diagnosis cannot obtain the collision frequency of plasma, and the laser itself has a high frequency, which has a great error in diagnosing plasma with low electron density, so laser diagnosis is not used in this work.

Microwave diagnosis is also a commonly used noncontact diagnosis method, which regards plasma as a medium with a constant dielectric constant. The principle of microwave diagnosis is simple, with little interference to plasma, low cost, and wide application. Microwave diagnosis can be divided into reflection diagnosis and transmission diagnosis [21–23]. Microwave reflection diagnosis can calculate the electron density at different positions of the plasma based on the reflection phase, so as to diagnose the spatial distribution of the electron density. Microwave transmission diagnosis can diagnose electron density and collision frequency based on the attenuation and phase shift caused by plasma absorption when electromagnetic waves are transmitted in plasma.

In this work, microwave transmission diagnosis is used in the diagnosis of ICP and the diagnosis error is analyzed. ICP has the characteristics of long duration, high temperature, high electron density and uniform parameter distribution. During the transmission of electromagnetic waves in the plasma, in addition to the attenuation caused by absorption by the plasma, there is also attenuation caused by reflection and diffraction, which leads to large errors in the diagnosis of the collision frequency of the plasma, making the diagnosis difficult. Therefore, a curve fitting method for the attenuation of distortion is proposed to reduce the attenuation caused by reflection, so as to calibrate the collision frequency for diagnosis.

The paper is arranged as follows. In section 2, the theory of microwave transmission method diagnosis and the theoretical basis of the curve fitting method are introduced. In section 3, the curve fitting method is introduced for parameter diagnosis of plasma with different electron densities and collision frequencies. The description of the microwave diagnostic system and the diagnosis of the measured attenuation are given in section 4. The conclusion is given in section 5.

2. Theoretical analysis

2.1 Theoretical analysis of plasma microwave transmission diagnosis

Plasma is considered to be a conductive dispersive medium with uniform dielectric constant in the microwave transmission diagnosis. The electrodynamic equation is used to describe the movement of charged particles in the plasma, and the dielectric constant and conductivity of the plasma are derived.

Combining the electrodynamic equation and the parameter characteristics of the plasma, the equivalent dielectric constant of the plasma can be calculated as:

${\tilde{ε}}_{r} = 1 - \frac{ω_{p}^{2}}{ω (ω - j v_{e})}$

(1)

where ω is the angular frequency of the incident electromagnetic wave, v_e is the plasma collision frequency, ω_p is the characteristic angular frequency of the plasma, and the expression of ω_p is as follows:

$ω_{p} = \sqrt{\frac{n_{e} e^{2}}{m_{e} ε_{0}}} .$

(2)

According to Maxwell's electromagnetic equations, the wave number $\tilde{k}$ and the equivalent complex permittivity ${\tilde{ε}}_{r}$ when the electromagnetic wave propagates in the plasma satisfy:

$\tilde{k} = β - j α = ω \sqrt{μ ε} = ω \sqrt{μ_{0} ε_{0} {\tilde{ε}}_{r}} = \frac{ω}{c} \sqrt{{\tilde{ε}}_{r}} .$

(3)

Therefore, the attenuation constant α and phase shift constant β of the electromagnetic wave can be obtained as:

$\begin{matrix} α = \frac{ω}{\sqrt{2} c} [(\frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}} - 1) \\ + {\sqrt{{(1 - \frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}})}^{2} + {(\frac{v_{e}}{ω} \cdot \frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}})}^{2}}]}^{\frac{1}{2}} \end{matrix}$

(4)

$\begin{matrix} β = \frac{ω}{\sqrt{2} c} [(1 - \frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}}) \\ + {\sqrt{{(1 - \frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}})}^{2} + {(\frac{v_{e}}{ω} \cdot \frac{ω_{p}^{2}}{ω^{2} + {v_{e}}^{2}})}^{2}}]}^{\frac{1}{2}} \end{matrix}$

(5)

where c is the velocity of light in the vacuum. According to the equations (1)–(5), the expressions of electron density and collision frequency are as follows:

$n_{e} = \frac{ε_{0} m_{e} [4 α^{2} β^{2} c^{4} + {(c^{2} (α^{2} - β^{2}) + ω^{2})}^{2}]}{e^{2} [c^{2} (α^{2} - β^{2}) + ω^{2}]}$

(6)

$ν_{e} = \frac{2 α β ω c}{c^{2} (α^{2} - β^{2}) + ω^{2}} (rad)$

(7)

where m_e is the electronic mass, e is the electronic charge. After the electromagnetic wave passes through the plasma of thickness d, the attenuation Att and the phase shift ϕ are respectively:

$A t t = 20 \log_{10} (e^{- α d}) = - 8.69 α d (dB)$

(8)

$ϕ = - ϕ_{p} - (- ϕ_{0}) = - (β - β_{0}) d = - (β - \frac{ω}{c}) d$

(9)

where ϕ_p is the phase in which the electromagnetic wave propagates in the plasma, ϕ₀ is the phase in which the electromagnetic wave propagates in free space, β₀ is the phase constant of the electromagnetic wave in the free space. It can be seen from equations (8) and (9) that the attenuation and phase shift of electromagnetic waves are related to electron density n_e, collision frequency ν_e, plasma thickness d, and incident electromagnetic wave angular frequency ω. It should be noted that both attenuation and phase shift are relative to the case without plasma. The vector network measurement obtains the transmission coefficient S₂₁ and the reflection coefficient S₁₁, then Att=the amplitude of S₂₁ (with plasma) minus the amplitude of S₂₁ (without plasma). Φ=S₂₁ phase (with plasma) minus S₂₁ phase (without plasma). The reflection attenuation and phase shift can be obtained in the same way.

As a consequence, the electron density and collision frequency can be diagnosed by measuring the attenuation and phase shift of electromagnetic waves as they propagate in the plasma with a vector net. In the Computer Simulation Technology (CST) software, the plasma model has a set electron density n_e0 and collision frequency ν_e0. Through the simulation of the diagnosis system, the attenuation and phase shift of electromagnetic wave transmission in the plasma can be obtained, and the diagnosed electron density n_e1 and collision frequency ν_e1 can be obtained. Comparing the two parameters, the diagnosis error can be obtained.

For plasmas with different electron densities, the attenuation of electromagnetic waves transmitted in them is shown in figure 1.

Figure 1. Attenuation of electromagnetic waves at different electron densities.

DownLoad: Full-Size Img PowerPoint

It can be seen from figure 1 that when the frequency of the incident electromagnetic wave is close to the characteristic frequency of the plasma, the attenuation increases sharply, indicating that the microwave transmission diagnosis can only diagnose the plasma with the characteristic frequency lower than the incident frequency.

In 2021, Chengwei Zhao studied the diagnostic accuracy of microwave transmission method [24]. In section 4.3 of that article, the parameters of plasma are diagnosed by simulation, and the diagnosis accuracy is given. The diagnostic accuracy of electron density is more than 90%, while the diagnostic accuracy of collision frequency is much lower than that of electron density. Since the simulation environment of the two articles is the same, it can be considered that the conclusion is also applicable to this paper.

2.2 Relationship between attenuation and collision frequency

$\begin{matrix} \frac{\partial α}{\partial v_{e}} = - \frac{\frac{2 ω v_{e} {ω_{p}}^{2}}{{(ω^{2} + {v_{e}}^{2})}^{2}} + \frac{\frac{4 ω v_{e} {ω_{p}}^{2} (\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}{{(ω^{2} + {v_{e}}^{2})}^{2}} + \frac{4 {v_{e}}^{3} {ω_{p}}^{4}}{ω {(ω^{2} + {v_{e}}^{2})}^{3}} - \frac{2 v_{e} {ω_{p}}^{4}}{ω {(ω^{2} + {v_{e}}^{2})}^{2}}}{2 \sqrt{{(\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}^{2} + \frac{{v_{e}}^{2} {ω_{p}}^{4}}{ω^{2} {(ω^{2} + {v_{e}}^{2})}^{2}}}}}{2 \sqrt{\sqrt{{(\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}^{2} + \frac{{v_{e}}^{2} {ω_{p}}^{4}}{ω^{2} {(ω^{2} + {v_{e}}^{2})}^{2}}} + \frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1}} \end{matrix}$

(10)

$\begin{matrix} \frac{\partial β}{\partial v_{e}} = \frac{\frac{2 ω v_{e} {ω_{p}}^{2}}{{(ω^{2} + {v_{e}}^{2})}^{2}} - \frac{\frac{4 ω v_{e} {ω_{p}}^{2} (\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}{{(ω^{2} + {v_{e}}^{2})}^{2}} + \frac{4 {v_{e}}^{3} {ω_{p}}^{4}}{ω {(ω^{2} + {v_{e}}^{2})}^{3}} - \frac{2 v_{e} {ω_{p}}^{4}}{ω {(ω^{2} + {v_{e}}^{2})}^{2}}}{2 \sqrt{{(\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}^{2} + \frac{{v_{e}}^{2} {ω_{p}}^{4}}{ω^{2} {(ω^{2} + {v_{e}}^{2})}^{2}}}}}{2 \sqrt{\sqrt{{(\frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} - 1)}^{2} + \frac{{v_{e}}^{2} {ω_{p}}^{4}}{ω^{2} {(ω^{2} + {v_{e}}^{2})}^{2}}} - \frac{{ω_{p}}^{2}}{ω^{2} + {v_{e}}^{2}} + 1}} \end{matrix}$

(11)

Set the electron density n_e=1.0×10¹⁹ m^-3, and the collision frequency ν_e=1.5 GHz. The relationship between the partial derivative of the attenuation constant α and the phase shift constant β to the collision frequency and the normalized frequency is shown in figure 2.

Figure 2. Partial derivative of α and β to the collision frequency.

DownLoad: Full-Size Img PowerPoint

It can be seen from figure 2 that when the incident frequency is greater than the characteristic frequency, the attenuation is mainly related to the collision frequency. Therefore, the following conclusions can be drawn: the error of the collision frequency diagnosis is mainly caused by the attenuation distortion, and the key to calibrating the collision frequency of the diagnosis is to calibrate the attenuation.

3. Calibration of collision frequency

Because part of the electromagnetic wave is reflected or diffracted when propagating in the plasma, and is not absorbed by the plasma, and this part of the attenuation is still included in the measured attenuation, the attenuation is distorted. It is necessary to remove the attenuation of electromagnetic wave reflection and diffraction from the measured attenuation, and obtain the attenuation caused only by plasma absorption.

3.1 Calibration using focusing lens antennas

A focusing lens antenna with a working frequency of 26.5–40 GHz is used to reduce the diffraction of electromagnetic waves [25]. The focal length of the antenna is 350 mm, and the lens resolution is smaller than 30 mm. Since the dielectric material used in the lens is quartz, it can withstand a high temperature of 1000 ℃. The electric field intensity along the transmission direction of the electromagnetic wave was tested in a microwave anechoic chamber. The structure and focusing effect of the antenna are shown in figure 3. The simulated and measured electric field in CST is shown in figure 3(b). It can be seen from figure 3(b) that the antenna has a good focusing performance, and measurement and simulation are in good agreement.

DownLoad: Full-Size Img PowerPoint

3.2 Calibration using curve fitting method

The working frequency of the focusing lens antenna is 26–40 GHz, the electron density of plasma n_e=1.0×10¹⁹ m^-3, and the collision frequency ν_e=1.5 GHz, and the relationship between attenuation and normalized frequency is shown in figure 4.

Figure 4. Relationship between attenuation and normalized frequency.

DownLoad: Full-Size Img PowerPoint

It can be seen from figure 4 that the attenuation curve with frequency is in the form of a rational function.

When the frequency of the incident electromagnetic wave is close to the characteristic frequency ω_p of the plasma (ω_p < ω < 1.05ω_p), the attenuation changes extremely fast. During this time period, it is difficult for the electromagnetic wave to pass through the plasma. Almost all electromagnetic waves are reflected, and the attenuation caused by the reflection is very large. The amplitude of the reflection coefficient S₁₁ measured by the vector network can be approximated as the attenuation caused by the plasma reflection. As a consequence, the amplitude difference between S₂₁ and S₁₁ is the attenuation caused by plasma absorption in this frequency band.

When the frequency of the incident electromagnetic wave is high (ω > 1.2ω_p), the attenuation is basically unchanged. In this frequency band, the electromagnetic wave completely penetrates the plasma, and the part reflected by the plasma is approximately zero, so the amplitude of S₂₁ is the attenuation caused by plasma absorption.

As a result, the attenuation of these two frequency bands can be used as a reference, and curve fitting can be performed in the form of a rational function, so as to restore the plasma attenuation of electromagnetic wave transmission. Generally, in the frequency band near the characteristic frequency of the plasma, the selection standard of the curve fitting benchmark is that the attenuation is in the range of 10–40 dB, and the selection standard in the high frequency band is that the attenuation is less than 3 dB.

4. Analysis of diagnostic accuracy

The experiment on microwave transmission plasma is simulated in CST software, and the collision frequency of plasma is diagnosed with the obtained attenuation and phase shift. The above conclusions are verified by comparing the diagnosis results before and after treatment with the curve fitting method.

4.1 Diagnosis of plasmas with different electron densities

The calibration method can be divided into three types according to the different electron density of the plasma. The electron densities (n_e) of the plasma are 0.75×10¹⁹ m^-3, 1.0×10¹⁹ m^-3, 1.5×10¹⁹ m^-3 respectively, and the collision frequency is 1.5 GHz. In CST, the microwave transmission plasma experiments with electron densities of 0.75×10¹⁹ m^-3, 1.0×10¹⁹ m^-3, 1.5×10¹⁹ m^-3 were simulated respectively, and the simulated attenuation was obtained. The curve fitting method is used to calibrate the simulated attenuation to obtain the calibrated attenuation. At the same time, the theoretical attenuation calculated by equations (4) and (8) is used as the reference value, which is used as the standard to measure the attenuation error. Figure 5 shows the comparison of the calibrated attenuation with the theoretical attenuation and the simulated attenuation.

Figure 5. Calibration of the attenuation of plasma with electron density of (a) 0.75×10¹⁹ m^-3, (b) 1.0×10¹⁹ m^-3 and (c) 1.5×10¹⁹ m^-3.

DownLoad: Full-Size Img PowerPoint

(1) n_e=0.75×10¹⁹ m^-3. The corresponding characteristic frequency is 24.6 GHz, which is not in the diagnostic frequency band, so only the high frequency attenuation is used for curve fitting, and the attenuation function obtained by curve fitting is:

$A t t = \frac{- 2.062 f + 60.49}{f - 30.53} .$

(12)

The calibration effect of the curve fitting method is shown in figure 5(a).

(2) n_e=1.0×10¹⁹ m^-3. The corresponding characteristic frequency is 28.4 GHz, which is within the diagnostic frequency band. According to the theory proposed in section 3, the attenuation corresponding to the frequency close to the characteristic frequency of the plasma and the attenuation corresponding to the high frequency can be used for curve fitting. The attenuation function obtained by curve fitting is:

$A t t = \frac{- 22.5}{f - 27.9} .$

(13)

The calibration effect of the curve fitting method is shown in figure 5(b).

(3) n_e=1.5×10¹⁹ m^-3. The corresponding characteristic frequency is 34.8 GHz, which is within the diagnostic frequency band. According to the theory proposed in section 3, the attenuation corresponding to the frequency close to the characteristic frequency of the plasma and the attenuation corresponding to the high frequency can be used for curve fitting. The attenuation function obtained by curve fitting is:

$A t t = \frac{- 20.87}{f - 34.38} .$

(14)

The calibration effect of the curve fitting method is shown in figure 5(c).

It can be seen from figure 5 that compared with the measured attenuation, the attenuation calibrated by the curve fitting method is closer to the theoretical attenuation. In the high frequency bands, the theoretical attenuation is in good agreement with the calibrated attenuation. The calibration effect on the collision frequency is shown in table 1.

Table 1. Diagnostic calibration of plasma with electron densities of 0.75×10¹⁹ m^-3, 1.0×10¹⁹ m^-3 and 1.5×10¹⁹ m^-3.

Electron density (m^-3)	Incident frequency (GHz)	Theoretical attenuation (dB)	Set collision frequency (GHz)	Measured attenuation (dB)	Diagnosed collision frequency (GHz)	Calibrated attenuation (dB)	Calibrated collision frequency (GHz)
0.75×10¹⁹	28	-4.985 86	1.5	-7.913 52	2.01/33.69%	-4.978 21	1.26/-15.89%
	33	-2.783 92		-2.702 03	1.31/-12.44%	-2.712 65	1.35/-9.79%
	38	-1.775 45		-2.586 79	2.04/35.78%	-1.856 61	1.46/-2.54%
1.0×10¹⁹	28	-127.949		-63.1023	0.42/-71.70%	-142.405	1.23/-17.73%
	33	-4.695 31		-7.412 85	2.11/40.43%	-4.411 76	1.24/-17.19%
	38	-2.734 41		-4.269 15	2.19/46.15%	-2.225 08	1.14/-23.83%
1.5×10¹⁹	35	-15.6857		-49.0878	2.42/61.23%	-28.9861	1.43/-4.63%
	37	-5.615 84		-17.6812	2.87/91.35%	-7.953 51	1.29/-13.90%
	39	-3.812 86		-11.6989	2.89/92.65%	-4.511 46	1.11/-25.70%

| Show Table

DownLoad: CSV

Table 1 demonstrates that the calibrated attenuation matches the theoretical attenuation better than the measured attenuation, and the calibrated collision frequency matches the set collision frequency well.

4.2 Diagnosis of plasmas with different collision frequencies

The calibration effect of the curve fitting method on the attenuation is different depending on the collision frequency of the plasma. In this part, the calibration effect of the curve fitting method on the attenuation is studied when the collision frequencies of the plasma are 1 GHz, 3 GHz, and 5 GHz, respectively. Figure 6 shows the comparison of the calibrated attenuation with the theoretical attenuation and the simulated attenuation.

Figure 6. Calibration of the attenuation of plasma with collision frequencies of (a) 1 GHz, (b) 3 GHz and (c) 5 GHz.

DownLoad: Full-Size Img PowerPoint

(1) ν_e=1 GHz. The attenuation function obtained by curve fitting is:

$A t t = \frac{- 0.7515 f + 12.22}{f - 28.46} .$

(15)

The calibration effect of the curve fitting method is shown in figure 6(a).

(2) ν_e=3 GHz. The attenuation function obtained by curve fitting is:

$A t t = \frac{- 1.51 f + 1.723}{f - 27.7} .$

(16)

The calibration effect of the curve fitting method is shown in figure 6(b).

(3) ν_e=5 GHz. The attenuation function obtained by curve fitting is:

$A t t = \frac{- 2.515 f + 13.27}{f - 27.8} .$

(17)

The calibration effect of the curve fitting method is shown in figure 6(c).

It can be seen from figure 6 that compared with the measured attenuation, the attenuation calibrated by the curve fitting method is closer to the theoretical attenuation. The calibration effect on the collision frequency is shown in table 2.

Table 2. Diagnostic calibration of plasma with electron densities of 0.75×10¹⁹ m^-3, 1.0×10¹⁹ m^-3 and 1.5×10¹⁹ m^-3.

Collision frequency (GHz)	Incident frequency (GHz)	Theoretical attenuation (dB)	Set electron density (m^-3)	Measured attenuation (dB)	Diagnosed collision frequency (GHz)	Calibrated attenuation (dB)	Calibrated collision frequency (GHz)
1	29	-8.053 43	1.0×10¹⁹	-22.6510	2.19/118.78%	-11.2843	1.09/9.06%
	34	-2.751 56		-4.6529	1.52/51.83%	-2.4382	0.79/-20.44%
	39	-1.678 77		-2.8878	1.62/62.61%	-1.6415	0.92/-7.57%
3	29	-26.5066		-40.8230	3.56/18.67%	-28.6440	2.50/-16.62%
	34	-8.2506		-10.4375	3.41/13.55%	-7.8676	2.57/-14.40%
	39	-5.0351		-6.6559	3.76/25.19%	-5.0571	2.85/-4.88%
5	29	-48.3406		-58.0459	4.97/7.00%	-49.3307	4.23/-15.48%
	34	-13.7375		-16.0458	5.25/4.92%	-11.6398	3.81/-23.87%
	39	-8.3881		-9.9575	5.63/12.45%	-7.5701	4.27/-14.50%

| Show Table

DownLoad: CSV

It can be seen from table 2 that for the plasma whose collision frequency is in the range of 1–10 GHz, the curve fitting method has a good calibration effect on attenuation. As the collision frequency increases, the calibration effect becomes worse. Beyond an order of magnitude of the collision frequency range, there is almost no calibration effect.

5. Diagnosis test and analysis

5.1 Erection of the experimental device

In the plasma generating device, the plasma is generated by inductive coupling, and is ejected into the vacuum microwave anechoic chamber through a circular nozzle with a diameter of 200 mm [26, 27]. The plasma microwave transmission diagnosis system mainly includes a vector network analyzer, focusing lens antennas, coaxial cables, sealing flanges and antenna brackets. The specific connections between the various parts are shown in figure 7. The functions of each device are as follows. The vector network analyzer contains a microwave transmitter and a microwave receiver, which are responsible for receiving and sending microwave signals. The transceiver antenna radiates microwave signals into free space. The coaxial cable is used to connect the vector network and the transmitting and receiving antennas for the transmission of microwave signals. A coaxial adapter device is installed on the sealing flange to connect the transmission of microwave signals inside and outside the vacuum chamber. The antenna bracket supports the antenna and can adjust the height of the antenna.

Figure 7. Plasma microwave transmission diagnostic system.

DownLoad: Full-Size Img PowerPoint

The diagnostic frequency band is 26–40 GHz, the voltage is 10 kV, and the air pressure in the vacuum chamber is maintained at 300 Pa. The distance between the center of the lens and the nozzle is 160 mm, and the distance between the transmitting and receiving antennas is 1100 mm. The vector network analyzer is used to collect five sets of data (amplitude and phase of S₂₁) for the experiments with and without plasma respectively, the attenuation and phase shift are obtained, and the average value is taken. The state of the plasma is shown in figure 8.

Figure 8. Attenuation of electromagnetic waves at different electron densities.

DownLoad: Full-Size Img PowerPoint

5.2 Analysis of experimental data

The measured attenuation and the calibrated attenuation obtained from the experiment are shown in figure 9.

Figure 9. Measured attenuation and calibrated attenuation.

DownLoad: Full-Size Img PowerPoint

According to the measured phase shift, it can be diagnosed that the electron density of the plasma is 4.15×10¹⁸ m^-3, and the corresponding characteristic frequency is 18.3 GHz. It can be seen from figure 9 that the characteristic frequency is lower than the diagnostic frequency band, but the measured attenuation is similar to the case (2) in section 4.1. This is because the electron density distribution of ICP is 'high in the middle-low around the periphery', which is not uniformly distributed like the simulated plasma. When the electromagnetic wave propagates in the plasma, it has been cut off at the central plasma (with high electron density), but it can still be transmitted in the surrounding plasma (with low electron density), which causes the measurement attenuation similar to the case (2) in section 4.1.

At the same time, it can be seen from figure 9 that the measured attenuation curve has a lot of jitters. This is due to the fact that some electromagnetic waves are reflected and refracted at the lens, and electromagnetic waves transmitted in different directions are superimposed on each other. Multiple sets of data at different times are collected, and the average value is calculated, which can reduce jitter to a certain extent.

The calibrated attenuation is obtained through a curve fitting method. Due to the lack of research on the diagnosis of collision frequency, there is no diagnostic result that it can be compared with. As a consequence, a method for verifying the calibration effect of the curve fitting method is proposed. The measured attenuation and the calibrated attenuation are respectively used to diagnose the collision frequency of the plasma, which are set as ν_e0 and ν_e1, and the values are assigned to the plasma model of CST, and the microwave transmission plasma experiment is simulated to obtain Att₀ and Att₁. Comparing them with the measured attenuation, if Att₁ (corresponding to the attenuation after calibration) is closer to the measured attenuation, it means that the curve fitting method has a good calibration effect for the diagnosis of collision frequency. On the contrary, if Att₀ (corresponding to the uncalibrated attenuation) is closer to the measured attenuation, it means that the curve fitting method has no diagnostic effect. The comparison of Att₀, Att₁ and the measured attenuation is shown in figure 10.

Figure 10. Att₀, Att₁ and the measured attenuation.

DownLoad: Full-Size Img PowerPoint

It can be seen from figure 10 that compared to Att₀, Att₁ is more consistent with the measured attenuation, which shows that the curve fitting method has a good calibration effect for the diagnosis of collision frequency. Since the true value of the collision frequency cannot be obtained, the diagnosis accuracy of the collision frequency cannot be given directly. However, the attenuation error can be obtained from figure 10. Taking the measured attenuation as a standard, the error of Att₁ is in the range of 0%–30% with a mean of 15.9%, while the error of Att₀ is about 20%–80% with a mean of 44.0%. It can be seen from the diagnosis results that the actual collision frequency should be between 1 GHz and 3 GHz. Taking the simulation results in chapter 4 as a reference, the collision frequency accuracy diagnosed by Att₀ is about 40%–60%, while the collision frequency accuracy diagnosed by Att₁ is about 80%–90%; the curve fitting method can improve the diagnostic accuracy of collision frequency by about 40%.

6. Conclusion

In this work, the microwave transmission diagnosis is used to obtain the collision frequency of the plasma, and the error of the diagnosis is analyzed. As part of the electromagnetic waves are reflected or diffracted, the measurement attenuation is distorted, and the diagnosis error of the collision frequency is large. In order to solve this problem and reduce the diagnostic error of collision frequency, a curve fitting method is proposed. For plasmas with different electron densities and collision frequencies, the diagnosis errors of collision frequency, before and after calibration with the curve fitting method, were studied and compared. The results of simulation and experiment are used to verify the effectiveness of the curve fitting method. At the same time, experiments show that the curve fitting method is also suitable for plasma with a nonuniform distribution of electron density.

Acknowledgments

This work was supported by National Natural Science Foundation of China (Nos. 12275095, 11975174, 11905030, 12011530142), the National Magnetic Confinement Fusion Energy Research Project of China (No. 2017YFE0301805) and the Fundamental Research Funds for the Central Universities (No. 2020IB023).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References (109)

References

[1]	Raizer Y P and Allen J E 1997 Gas Discharge Physics vol 2 (Berlin: Springer)
[2]	Paschen F 1889 Ueber die Zum Funkenübergang in Luft: Wasserstoff und Kohlensäure Bei Verschiedenen Drucken Erforderliche Potentialdifferenz (Berlin: JA Barth)
[3]	Townsend J S 1915 Electricity in Gases (Oxford: Clarendon)
[4]	Pejovic M M, Ristic G S and Karamarkovic J P 2002 J. Phys. D: Appl. Phys. 35 R91 doi: 10.1088/0022-3727/35/10/201
[5]	Graf R F 1999 Modern Dictionary of Electronics (Woburn: Newnes)
[6]	Lieberman M A and Lichtenberg A J 2005 Principles of Plasma Discharges and Materials Processing (New York: Wiley)
[7]	Naidu M S 2013 High Voltage Engineering (New York: McGraw-Hill)
[8]	Raizer Y P and Allen J E 1991 Gas Discharge Physics vol 1 (Berlin: Springer)
[9]	Bogaerts A et al 2002 Spectrochim. Acta, Part B: Atomic Spectroscopy 57 609 doi: 10.1016/S0584-8547(01)00406-2
[10]	Boyle W and Kisliuk P 1955 Phys. Rev. 97 255 doi: 10.1103/PhysRev.97.255
[11]	Cookson A H 1970 Electrical breakdown for uniform fields in compressed gases Proc. of the Institution of Electrical Engineers (London: IET), vol 117, 269
[12]	Loveless A M and Garner A L 2017 Phys. Plasmas 24 104501 doi: 10.1063/1.4998245
[13]	Savic M et al 2011 IEEE Trans. Plasma Sci. 39 2556 doi: 10.1109/TPS.2011.2159244
[14]	Chabert P and Braithwaite N 2011 Physics of RadioFrequency Plasmas (Cambridge: Cambridge University Press)
[15]	Al Bastami A et al 2020 Comparison of radio-frequency power architectures for plasma generation 2020 IEEE XXI Workshop on Control and Modeling for Power Electronics (COMPEL) (Piscataway, NJ: IEEE) pp 1–8
[16]	Verma A K and Venkattraman A 2021 Comput. Phys. Commun. 263 107855 doi: 10.1016/j.cpc.2021.107855
[17]	Tolpygo S K et al 2016 IEEE Trans. Appl. Supercond. 26 1 doi: 10.1109/TASC.2016.2519388
[18]	Lieberman M A et al 2006 Plasma Sources Sci. Technol. 15 276 doi: 10.1088/0963-0252/15/2/013
[19]	Chang H W and Hsu C C 2010 Jpn. J. Appl. Phys. 49 016101 doi: 10.1143/JJAP.49.016101
[20]	Obermair C et al 2022 Phys. Rev. Accel. Beam. 25 104601 doi: 10.1103/PhysRevAccelBeams.25.104601
[21]	Su Z X et al 2021 Plasma Sources Sci. Technol. 30 125013 doi: 10.1088/1361-6595/ac3e3f
[22]	Donkó Z and Petrović Z L 2006 Japan. J. Appl. Phys. 45 8151 doi: 10.1143/JJAP.45.8151
[23]	Delattre P A et al 2013 J. Phys. D: Appl. Phys. 46 235201 doi: 10.1088/0022-3727/46/23/235201
[24]	Lafleur T and Chabert P 2015 Plasma Sources Sci. Technol. 24 044002 doi: 10.1088/0963-0252/24/4/044002
[25]	Yang S et al 2017 J. Vacuum Sci. Technol. A: Vacuum, Surfaces, and Films 35 061311 doi: 10.1116/1.4990137
[26]	Wu H et al 2022 Plasma Sources Sci. Technol. 31 047001 doi: 10.1088/1361-6595/ac5c60
[27]	Vender D, Smith H and Boswell R 1996 J. Appl. Phys. 80 4292 doi: 10.1063/1.363389
[28]	Kawamura E, Birdsall C K and Vahedi V 2000 Plasma Sources Sci. Technol. 9 413 doi: 10.1088/0963-0252/9/3/319
[29]	Wang H Y, Jiang W and Wang Y N 2010 Plasma Sources Sci. Technol. 19 045023 doi: 10.1088/0963-0252/19/4/045023
[30]	Lisovsky V and Yegorenkov V 1994 J. Phys. D: Appl. Phys. 27 2340 doi: 10.1088/0022-3727/27/11/014
[31]	Lisovskiy V and Yegorenkov V 1998 J. Phys. D: Appl. Phys. 31 3349 doi: 10.1088/0022-3727/31/23/008
[32]	Korolov I, Derzsi A and Donkó Z 2014 J. Phys. D: Appl. Phys. 47 475202 doi: 10.1088/0022-3727/47/47/475202
[33]	Kirchner F 1925 Annalen der Phys. 382 287 doi: 10.1002/andp.19253821105
[34]	Githens SJr 1940 Phys. Rev. 57 822 doi: 10.1103/PhysRev.57.822
[35]	Levitskii S 1957 Sov. Phys. Tech. Phys. 2 887
[36]	Kihara T 1952 Rev. Mod. Phys. 24 45 doi: 10.1103/RevModPhys.24.45
[37]	Lisovskiy V A and Yegorenkov V D 1999 J. Phys. D: Appl. Phys. 32 2645 doi: 10.1088/0022-3727/32/20/310
[38]	Walsh J L et al 2008 Appl. Phys. Lett. 93 221505 doi: 10.1063/1.3043449
[39]	Lisovskiy V, Kharchenko N and Yegorenkov V 2010 J. Phys. D: Appl. Phys. 43 425202 doi: 10.1088/0022-3727/43/42/425202
[40]	Dudin S et al 2012 Questions of Atomic Science and Technology (in Russian)
[41]	Dakhov A and Dudin S 2013 Questions of Atomic Science and Technology (in Russian)
[42]	Petrović Z L et al 2014 J. Phys. Conf. Ser. 514 012043 doi: 10.1088/1742-6596/514/1/012043
[43]	Lisovskiy V A, Derevianko V A and Yegorenkov V D 2015 J. Phys. D: Appl. Phys. 48 475201 doi: 10.1088/0022-3727/48/47/475201
[44]	Huo W et al 2018 IEEE Trans. Plasma Sci. 46 8 doi: 10.1109/TPS.2017.2780101
[45]	Aponte I A et al 2019 Phys. Plasmas 26 123512 doi: 10.1063/1.5119152
[46]	Liu Y X et al 2020 Plasma Sources Sci. Technol. 29 12LT03 doi: 10.1088/1361-6595/abcc7a
[47]	Wang X Y et al 2021 Plasma Sources Sci. Technol. 30 075011 doi: 10.1088/1361-6595/ac0b56
[48]	Hale D H 1948 Phys. Rev. 73 1046 doi: 10.1103/PhysRev.73.1046
[49]	Pim J 1949 Proc. of the IEE-Part Ⅲ: Radio and Communication Engineering 96 117 doi: 10.1049/pi-3.1949.0024
[50]	Pim J 1948 Nature 161 683 doi: 10.1038/161683b0
[51]	Dosopoulos S and Lee J F 2010 J. Comput. Phys. 229 8521 doi: 10.1016/j.jcp.2010.07.036
[52]	Verma A K and Venkattraman A 2021 Comput. Phys. Commun. 263 107855 doi: 10.1016/j.cpc.2021.107855
[53]	Kühn C and Groll R 2021 Comput. Phys. Commun. 262 107853 doi: 10.1016/j.cpc.2021.107853
[54]	Fu Y et al 2021 Phys. Rev. Appl. 16 054016 doi: 10.1103/PhysRevApplied.16.054016
[55]	Petrović Z L et al 2017 Plasma Processes Polym. 14 1600124 doi: 10.1002/ppap.201600124
[56]	Puač M et al 2018 Plasma Sources Sci. Technol. 27 075013 doi: 10.1088/1361-6595/aacc0c
[57]	Puač M, Ɖorđević A and Petrović Z L 2020 Eur. Phys. J. D 74 1 doi: 10.1140/epjd/e2020-100526-1
[58]	Qiu X et al 2022 IEEE Transactions on Plasma Science (Piscataway, NJ: IEEE) pp 1–8
[59]	Deng Y et al 2008 Phys. Plasmas 15 053507 doi: 10.1063/1.2918660
[60]	Gao S H, Wang X L and Zhang Y T 2021 Phys. Plasmas 28 113501 doi: 10.1063/5.0061546
[61]	Birdsall C K 1991 IEEE Trans. Plasma Sci. 19 65 doi: 10.1109/27.106800
[62]	Vahedi V et al 1993 Plasma Sources Sci. Technol. 2 261 doi: 10.1088/0963-0252/2/4/006
[63]	Vahedi V and Surendra M 1995 Comput. Phys. Commun. 87 179 doi: 10.1016/0010-4655(94)00171-W
[64]	Nanbu K 2000 IEEE Trans. Plasma Sci. 28 971 doi: 10.1109/27.887765
[65]	Wen D Q et al 2017 J. Phys. D: Appl. Phys. 50 495201 doi: 10.1088/1361-6463/aa9627
[66]	Smith H, Charles C and Boswell R 2003 Phys. Plasmas 10 875 doi: 10.1063/1.1531615
[67]	Radmilović-Radjenović M and Lee J 2005 Phys. Plasmas 12 063501 doi: 10.1063/1.1922267
[68]	Lee M U et al 2017 Plasma Sources Sci. Technol. 26 034003 doi: 10.1088/1361-6595/aa52a8
[69]	Wu H et al 2021 Plasma Sources Sci. Technol. 30 065029 doi: 10.1088/1361-6595/abff74
[70]	Wu H et al 2022 J. Phys. D: Appl. Phys. 55 255203 doi: 10.1088/1361-6463/ac5772
[71]	Verboncoeur J P et al 1993 J. Comput. Phys. 104 321 doi: 10.1006/jcph.1993.1034
[72]	Vahedi V and DiPeso G 1997 J. Comput. Phys. 131 149 doi: 10.1006/jcph.1996.5591
[73]	Jiang M et al 2020 Plasma Sources Sci. Technol. 29 015020 doi: 10.1088/1361-6595/ab6755
[74]	Rauf S and Kushner M J 1998 J. Appl. Phys. 83 5087 doi: 10.1063/1.367326
[75]	Schmidt F, Mussenbrock T and Trieschmann J 2018 Plasma Sources Sci. Technol. 27 105017 doi: 10.1088/1361-6595/aae429
[76]	Schmidt F et al 2019 J. Appl. Phys. 125 173106 doi: 10.1063/1.5091965
[77]	Qu C et al 2020 J. Appl. Phys. 127 133302 doi: 10.1063/5.0002522
[78]	Wu H et al 2022 J. Appl. Phys. 131 153301 doi: 10.1063/5.0085311
[79]	Theis J G et al 2021 Phys. Plasmas 28 063513 doi: 10.1063/5.0051095
[80]	Xiao D 2016 Gas Discharge and Gas Insulation (Berlin: Springer)
[81]	Jauberteau J et al 1991 J. Phys. D: Appl. Phys. 24 261 doi: 10.1088/0022-3727/24/3/006
[82]	Haverlag M et al 1991 J. Appl. Phys. 70 3472 doi: 10.1063/1.350346
[83]	Rauf S and Kushner M J 1997 J. Appl. Phys. 82 2805 doi: 10.1063/1.366111
[84]	Booth J et al 1999 J. Appl. Phys. 85 3097 doi: 10.1063/1.369649
[85]	Gogolides E, Stathakopoulos M and Boudouvis A 1994 J. Phys. D: Appl. Phys. 27 1878 doi: 10.1088/0022-3727/27/9/011
[86]	Denpoh K and Nanbu K 1998 J. Vacuum Sci. Technol. A: Vacuum, Surfaces, and Films 16 1201 doi: 10.1116/1.581259
[87]	Denpoh K and Nanbu K 2000 Jpn. J. Appl. Phys. 39 2804 doi: 10.1143/JJAP.39.2804
[88]	Proshina O et al 2010 Plasma Sources Sci. Technol. 19 065013 doi: 10.1088/0963-0252/19/6/065013
[89]	Schulze J et al 2011 Phys. Rev. Lett. 107 275001 doi: 10.1103/PhysRevLett.107.275001
[90]	Liu G H et al 2015 Plasma Sources Sci. Technol. 24 034006 doi: 10.1088/0963-0252/24/3/034006
[91]	Liu Y X et al 2019 Plasma Sources Sci. Technol. 28 075005 doi: 10.1088/1361-6595/ab27aa
[92]	Jin C et al 2015 Plasma Processes Polym. 12 1061 doi: 10.1002/ppap.201400181
[93]	Georgieva V, Bogaerts A and Gijbels R 2004 Phys. Rev. E 69 026406 doi: 10.1103/PhysRevE.69.026406
[94]	Georgieva V and Bogaerts A 2005 J. Appl. Phys. 98 023308 doi: 10.1063/1.1989439
[95]	Kushner M J 2009 J. Phys. D: Appl. Phys. 42 194013 doi: 10.1088/0022-3727/42/19/194013
[96]	Lee C et al 1994 J. Electrochem. Soc. 141 1546 doi: 10.1149/1.2054960
[97]	Lee C and Lieberman M 1995 J. Vacuum Sci. Technol. A: Vacuum, Surfaces, and Films 13 368 doi: 10.1116/1.579366
[98]	Lloyd B, Carolan P and Warrick C 1996 Plasma Phys. Controlled Fusion 38 1627 doi: 10.1088/0741-3335/38/9/007
[99]	Kim H T 2013 Physics and computational simulations of plasma burn-through for tokamak start-up Ph. D. Thesis Imperial College London
[100]	Kim H T et al 2013 Plasma Phys. Controlled Fusion 55 124032 doi: 10.1088/0741-3335/55/12/124032
[101]	Hada K et al 2015 Fusion Sci. Technol. 67 693 doi: 10.13182/FST14-811
[102]	Horváth B et al 2017 Plasma Sources Sci. Technol. 26 124001 doi: 10.1088/1361-6595/aa963d
[103]	Lin M C, Zhou C and Smithe D N 2014 IEEE Trans. Electron Devices 61 1742 doi: 10.1109/TED.2014.2304715
[104]	Schmidt F, Mussenbrock T and Trieschmann J 2018 Plasma Sources Sci. Technol. 27 105017 doi: 10.1088/1361-6595/aae429
[105]	Yu S et al 2023 Comput. Phys. Commun. 282 108468 doi: 10.1016/j.cpc.2022.108468
[106]	Liu J K et al 2022 Phys. Plasmas 29 013508 doi: 10.1063/5.0075305
[107]	Chabert P 2007 J. Phys. D: Appl. Phys. 40 R63 doi: 10.1088/0022-3727/40/3/R01
[108]	Chabert P et al 2004 Phys. Plasmas 11 1775 doi: 10.1063/1.1688334
[109]	Chabert P et al 2004 Phys. Plasmas 11 4081 doi: 10.1063/1.1770900

[1]	Min ZHU (朱敏), Chao YE (叶超), Xiangying WANG (王响英), Amin JIANG (蒋阿敏), Su ZHANG (张苏). Effect of radio-frequency substrate bias on ion properties and sputtering behavior of 2 MHz magnetron sputtering[J]. Plasma Science and Technology, 2019, 21(1): 15507-015507. DOI: 10.1088/2058-6272/aae7dd
[2]	Amin JIANG (蒋阿敏), Chao YE (叶超), Xiangying WANG (王响英), Min ZHU (朱敏), Su ZHANG (张苏). Ion property and electrical characteristics of 60 MHz very-high-frequency magnetron discharge at low pressure[J]. Plasma Science and Technology, 2018, 20(10): 105401. DOI: 10.1088/2058-6272/aad379
[3]	Yan YUAN (袁燕), Lizhen YANG (杨丽珍), Zhongwei LIU (刘忠伟), Qiang CHEN (陈强). High power impulse magnetron sputtering and its applications[J]. Plasma Science and Technology, 2018, 20(6): 65501-065501. DOI: 10.1088/2058-6272/aa9e48
[4]	Yue HUA (滑跃), Jian SONG (宋健), Zeyu HAO (郝泽宇), Gailing ZHANG (张改玲), Chunsheng REN (任春生). Effects of direct current discharge on the spatial distribution of cylindrical inductivelycoupled plasma at different gas pressures[J]. Plasma Science and Technology, 2018, 20(1): 14005-014005. DOI: 10.1088/2058-6272/aa8ea8
[5]	Peifang YANG (杨培芳), Chao YE (叶超), Xiangying WANG (王响英), Jiamin GUO (郭佳敏), Su ZHANG (张苏). Control of growth and structure of Ag films by the driving frequency of magnetron sputtering[J]. Plasma Science and Technology, 2017, 19(8): 85504-085504. DOI: 10.1088/2058-6272/aa6619
[6]	Jiamin GUO (郭佳敏), Chao YE (叶超), Xiangying WANG (王响英), Peifang YANG (杨培芳), Su ZHANG (张苏). Growth and structural properties of silicon on Ag films prepared by 40.68 MHz veryhigh-frequency magnetron sputtering[J]. Plasma Science and Technology, 2017, 19(7): 75502-075502. DOI: 10.1088/2058-6272/aa6395
[7]	Liang SONG (宋亮), Xianping WANG (王先平), Le WANG (王乐), Ying ZHANG (张营), Wang LIU (刘旺), Weibing JIANG (蒋卫斌), Tao ZHANG (张涛), Qianfeng FANG (方前锋), Changsong LIU (刘长松). Fabrication and characterization of He-charged ODS-FeCrNi films deposited by a radio-frequency plasma magnetron sputtering technique[J]. Plasma Science and Technology, 2017, 19(4): 45502-045502. DOI: 10.1088/2058-6272/aa57f0
[8]	MIAO Chunguang (苗春光), WANG Xiangqin (王相勤). Mass Deposition, Etching and Sputtering Effects of Low-Energy N ⁺ Ion Irradiation on Solid Fly Ash[J]. Plasma Science and Technology, 2013, 15(12): 1232-1236. DOI: 10.1088/1009-0630/15/12/13
[9]	WANG Qing (王庆), WANG Yongfu (王永富), BA Dechun (巴德纯), YUE Xiangji (岳向吉). The Effect of Ion Current Density on Target Etching in Radio Frequency-Magnetron Sputtering Process[J]. Plasma Science and Technology, 2012, 14(3): 235-239. DOI: 10.1088/1009-0630/14/3/09
[10]	MU Zongxin, LIU Shengguang, ZANG Hairong, WANG Chun, MU Xiaodong. Discharge Properties of High-Power Pulsed Unbalanced Magnetron Sputtering[J]. Plasma Science and Technology, 2011, 13(6): 667-671.

Cited By

Cited by

Periodical cited type(2)

1.	Fu, Y., Ji, P., He, M. et al. Study of Plasma Particle Distribution and Electron Temperature in Cylindrical Magnetron Sputtering. Plasma Chemistry and Plasma Processing, 2023. DOI:10.1007/s11090-023-10425-9
2.	Fu, Y., Peng, J., He, M. et al. DC Magnetron Sputtering Particle Distribution and Energy Simulation Study. 2023. DOI:10.1109/ICEMPE57831.2023.10139580