ECR Layer Research Articles

Numerical simulation of magnetic field and channel configuration effect on cusped field thruster with electron cyclotron resonance discharge enhancement

Cusped field thrusters with electron cyclotron resonance discharge enhancement have great potential in micro-Newton propulsion field. According to the magnetic field and channel configuration, the thruster can be divided into 1-stage thruster, 2-stage thruster and 3-stage thruster. The ionization region and acceleration region of the 1-stage thruster are located at the channel upstream, which causes a large amount of wall power deposition, resulting in poor performance. Due to the shorter channel of the 2-stage thruster, it is easier for electrons to reach the ECR layer on the near-antenna region to obtain energy to promote ionization, compared with 3-stage thruster. In addition, the magnetic cusp number of the 2-stage thruster wall is lower than that of the 3-stage thruster, reducing the total wall power deposition. It is necessary to use 2-stage short channel configuration to design the thruster.

Effects of gas adsorbed on solid surface during gas breakdown in electron cyclotron resonance discharges

The microwave breakdown power (P wb) in an ECR plasma source was not merely determined by pressure (gas flow rate), but found to vary with the time interval between two successive breakdowns. The measured P wb dropped rapidly from a high value at a short time interval to a low level at a long time interval. The obtained dependence of P wb on pressure (gas flow rate) exhibited distinct features: the normal monotonicity and abnormal non-monotonicity at the short and long time intervals, respectively. The effective zone in the antenna’s surface, bombarded by hot electrons heated in the ECR layer, was validated by (1) masking the antenna with a film having a variable radius; (2) calculating the distribution of the vertical component of the microwave electric field with respect to the static magnetic field; and (3) imaging glows of transient breakdown discharges with a fast camera. The reduction in P wb was mainly attributed to the enhanced emission of δ-electrons from the gas-adsorbed antenna under the bombardment of energetic electrons coming from the ECR layer. The correlation between the dynamic gas coverage and the emission coefficient of δ-electrons was established to understand the abnormal ECR breakdown features.

Numerical simulation of influence of magnetic field on plasma characteristics and surface current of ion source of 2-cm electron cyclotron resonance ion thruster

The cathode-less miniature electron cyclotron resonance ion thruster (ECRIT) has the advantages of long-life and simple-structure. In the ECRIT ion source, the plasma distribution will affect the beam extraction, and the relative position of the ECR layer determined by the magnetic field structure and the flat-ring antenna together affect the plasma distribution. Due to the sheath, the ions or electrons in the plasma will be accelerated to sputter the surface of wall and induce plasma loss. It is important to investigate the wall currents and plasma characteristics. Therefore, particle-in-cell with Monte Carlo collision (PIC/MCC) model is established in this article to study the influence of the magnetic field structure on the plasma and wall current characteristics of 2-cm ECRIT ion source. The calculation results show that the electrons are confined near the ECR layer of antenna by the magnetic mirror, which leads the plasma to be distributed near the ECR layer. When the ECR layer is located on the upstream side of the flat-ring antenna, the plasma is concentrated between the antenna and magnet rings, and the ion density in front of the grid is lowest, which results in a lower ion beam current extracted from ion source and a lower current on the surface of magnetic ring and antenna. When the ECR layer is located on the downstream side of the flat-ring antenna, the plasma density near the upstream side of the antenna and grid is high, which results in higher ion beam current extracted from the ion source and higher current on the surface of antenna and magnetic ring. The plasma distribution and the total wall current of the ion source are affected weakly by the magnetic field structure. In this magnetic field structure, the ion sputtering on the flat-ring antenna is serious. Although such a magnetic field design can increase the extracted ion beam current, it will shorten the working life of the ion source. In the future, when designing a new thruster, it is necessary to weigh the ion current of extraction and lifetime to select the appropriate magnetic field structure.

The Effect of Removing the High Power Pulse Transformer on the Microwave Pulse Duration in the Preionization Phase in Taban Tokamak

In this study, a simple, economical and efficient preionization system (2.45 GHz, 1 kW) was constructed to assist the current startup in Taban tokamak ( $B_{T}= 0.7$ T, $R = 0.45$ m, a = 0.15 m). Power pulse transformer is a part of the preionization circuit, and its removal from the system can have impact on the performance of microwave duration and plasma preionization. In the present system, the magnetron was connected directly to the capacitor using a triggered spark gap. This new preionization system allows studying the electron cyclotron emission (ECR) preionization phenomenon at different toroidal magnetic fields. The toroidal magnetic field ripple was also used to determine the entrance moment of the ECR layer into the chamber. As a result, the exact moment of sending microwave can be determined using the moment of the ECR layer formation inside the chamber for maximum absorption of wave power.

Low-power-threshold parametric decay instabilities of powerful microwave beams in toroidal fusion devices

We discuss the experimental conditions responsible for a drastic decrease in the power threshold of parametric decay instabilities under auxiliary electron cyclotron resonance heating (ECRH) in toroidal magnetic fusion devices when the upper hybrid (UH) resonance for the pump wave is absent. We show that for a finite-width pump in the presence of a nonmonotonic (hollow) density profile occurring due to plasma equilibrium in the magnetic islands or anomalous particle fluxes from the ECR layer, 3D localization of one or both daughter waves is possible. This localization leads to the full suppression of daughter wave energy losses from the decay layer and a substantial increase in the nonlinear pumping efficiency. This decreases the power threshold of nonlinear excitation, which can be easily overcome in current ECRH experiments utilizing 1 MW microwave beams. Different scenarios of extraordinary and ordinary wave decays are investigated. The secondary decays of primary daughter waves and pump wave depletion are considered as the most effective mechanisms leading to the transition of primary instability to the saturation regime. The proposed theoretical model was shown to be able to describe the anomalous phenomena discovered in ECRH experiments in different toroidal fusion devices all over the world.

Non-inductive build-up of a hot plasma by X-wave ECH/ECCD to remove initial high loop-voltage from start-up of fusion tokamaks

The possibility of non-inductive build-up of a hot plasma by X-wave ECH/ECCD to remove initial high loop-voltage stage from start-up of fusion tokamaks is theoretically investigated, using ITER parameters. There are two injection methods for X-wave ECH/ECCD in ITER start-up plasmas. The first is oblique X-wave injection from the low-field side, referred to as the DX method. For the low density start-up plasma at , the wave with GHz and is damped away via cyclotron resonance absorption before encountering the cyclotron cutoff layer when the electron temperature is as high as eV. The wave power is deposited into the resonance electrons of which parallel velocity is as fast as , which enables an efficient ECCD. The second method is oblique O-wave injection from the low-field side, referred to as the OX method. The waves pass through the ECR layer with negligible power deposition when eV and are mode-converted into oblique X-waves upon reflection on the inboard wall. The X-waves propagate toward the ECR layer without cutoff, and are cyclotron-damped away even when the temperature is as low as eV. The OX method is appropriate for the initial open field stage and transition stage of the discharge, while the DX method is appropriate for the final tokamak stage, where Te is high. Discussions are given on the choice of appropriate BV and location of ECR layer for initiation of closed flux surfaces. Nine model plasmas from an initial open field plasma to a final hot tokamak plasma, consisting of a series of snapshots of a build-up discharge, are generated using an equilibrium theory, enabling estimation of required RF power to drive the current. These investigations suggest that an injection power MW could build up a hot plasma with kA, , keV, m and m.

Non inductive formation of an extremely overdense spherical Tokamak by electron Bernstein wave heating and current drive on LATE

An extremely overdense special Tokamak plasma has been non-inductively formed and maintained by electron Bernstein (EB) wave heating and current drive in the Low Aspect ratio Torus Experiment (LATE) device. The plasma current reaches 12 kA and the line-averaged electron density exceeds 7 times the plasma cut off density by injecting a 2.45 GHz microwave power of 60 kW. Such a highly overdense plasma is obtained when the upper hybrid resonance layer lies to the higher field side of the 2nd harmonic ECR layer, which may realize a good coupling to EB waves at their first propagation band. The effect of the injection polarization on the mode conversion rate to EB waves at the extremely overdense regime has been investigated and an improvement in the plasma current is observed.

Three-dimensional particle-in-cell simulation of a miniature plasma source for a microwave discharge ion thruster

We have developed a three-dimensional particle model for a miniature microwave discharge ion thruster to elucidate the mechanism of ECR discharges confined in a small space. The model consists of a particle-in-cell simulation with a Monte Carlo collision algorithm (PIC-MCC) for the kinetics of charged particles, a finite-difference time-domain method for the electromagnetic fields of 4.2 GHz microwaves, and a finite element analysis for the magnetostatic fields of permanent magnets. The PIC-MCC results have shown that the electrons are well confined owing to the mirror magnetic fields and can be effectively heated in the ECR layer downstream of a ring-shaped antenna. The confinement results in the ring-shaped profiles of the plasma density along the antenna. The visual appearance of the plasma discharge of the thruster in operation was also ring-shaped. Moreover, the ions are expected to be accelerated effectively through the grid electrode without a large loss of ions toward side walls, that is, the plasma source developed here would be desirable in ion thrusters.

Full-wave FEM simulations of electromagnetic waves in strongly magnetized non-homogeneous plasma

Electromagnetic wave propagation in anisotropic media has been widely studied over the last decades since there are several applications where anisotropy plays an important role. This paper presents a procedure to carry out three-dimensional (3D) finite element method (FEM) full-wave simulations of the electromagnetic field in inhomogeneous magnetized plasma of an electron cyclotron resonance ion source (ECRIS). We used COMSOL Multiphysics software and MATLAB to model a cold anisotropic magnetized plasma, described by full-3D non-uniform dielectric tensor, enclosed by the metallic cylindrical cavity where the plasma is formed. A proper mesh generation, exploiting FEM-based COMSOL versatility, allowed us to optimally model ECRIS cavity and microwave waveguide launching structure, with a good computational efficiency and high resolution of the solution especially around the resonance regions. Numerical simulations have been performed in the frequency domain: in the resonance regions especially, the material properties exhibit a spatial variation that leads to a large sparse ill-conditioned matrix which is solved by MUltifrontal Massively Parallel Solver (MUMPS) direct method. We implement a method to perform full-wave simulations considering a cold plasma model for the constitutive relations; the obtained results show that the presence of ECR layer, along with the cavity walls, strongly influences the shape and strength of the electromagnetic field distribution, featuring a strong non-uniformity of the main electromagnetic parameters.

A double-layer based model of ion confinement in electron cyclotron resonance ion source

The paper proposes a new model of ion confinement in ECRIS, which can be easily generalized to any magnetic configuration characterized by closed magnetic surfaces. Traditionally, ion confinement in B-min configurations is ascribed to a negative potential dip due to superhot electrons, adiabatically confined by the magneto-static field. However, kinetic simulations including RF heating affected by cavity modes structures indicate that high energy electrons populate just a thin slab overlapping the ECR layer, while their density drops down of more than one order of magnitude outside. Ions, instead, diffuse across the electron layer due to their high collisionality. This is the proper physical condition to establish a double-layer (DL) configuration which self-consistently originates a potential barrier; this "barrier" confines the ions inside the plasma core surrounded by the ECR surface. The paper will describe a simplified ion confinement model based on plasma density non-homogeneity and DL formation.

On correlation measurements of electron Bernstein wave emission

A systematic study of the electron Bernstein wave (EBW) emission from either overdense plasmas or plasmas with moderate electron temperatures is presented. Starting from description of the bare electrons microscopic thermal radiation we finally arrive at the macroscopic expression for the radiation temperature. The latter is shown to be a sum of the electron temperature and a fluctuating piece caused by the fluctuations of both the EBW emission source and the EBW damping rate within the ECR layer. It is shown that the correlation measurements of the EBW emission at two separate frequencies reveal information on the correlation length of the low frequency turbulence.

Transient process of a spherical tokamak plasma startup by electron cyclotron waves

Non-inductive startup by a 2.45 GHz microwave through electron cyclotron resonance heating (ECRH) has been studied in the SUNIST spherical tokamak (R/a: 0.3 m/0.23 m; BT0: < 0.15 T). Two types of discharges, with quite different transient characteristics, were found. The transient processes of discharges are experimentally investigated by scanning the radial position of the ECR layer, the vertical magnetic field and the microwave power. Time domain analysis of microwave reflections and Hα emissions reveal a process dominated by the combination of ionization, loss along the open field lines and the gradient B drift. It is shown that the transient character is insensitive to the experimental parameters except the pressure of filled gas. This result confirms that the control of gas puffing is of great importance for starting up a spherical tokamak by ECRH using low-frequency microwaves.

Study of Spatial Distribution of Microwave Power Deposition for Plug ECRH in GAMMA 10

Wave propagation around the ECR layer in the GAMMA 10 plug region is analyzed by calculating the dispersion relation of an electron cyclotron wave in a hot plasma. Then, the spatial distribution of microwave power deposition and the absorption rate along each microwave ray are calculated. The absorption rate is experimentally evaluated by using an array of waveguide antennas. The calculated value well agrees with the experimental one on reasonable assumption that the extraordinary mode shares about 90% of the injected microwave power. This analysis is used to obtain an axisymmetric power deposition distribution. It is shown that the heating wave should be directed somewhat upward than the direction to the on-axis point on the resonance layer. This is because a larger power is deposited in the injection side lower side to the machine axis. For the plug in GAMMA 10, an injection beam with an elliptic cross section is suitable to obtain a circular distribution of power deposition.

Charge-exchange neutral particle measurements on electron cyclotron resonance heating experiments in hot-ion-mode plasmas of the GAMMA 10 tandem mirror

Charge-exchange (C–X) neutral particle measurements have been carried out in hot-ion-mode plasmas of the GAMMA 10 tandem mirror. In the present experiment, a microwave power of 40 kW in 28 GHz is injected toward a second harmonic ECR layer located in the vicinity of the ICR layer at the central region and the radial profiles of ion temperatures determined from the energy spectrum of the C–X neutrals by using a neutral particle energy analyzer (NPA) are investigated from the viewpoint of ion energy balance. At the onset of the ECRH pulse, a remarkable increase of C–X neutral flux with high energy (few keV to few tens keV) is observed with NPA and the resultant ion temperature on the plasma axis is found to increase from 2.5 to 5.0 keV at the electron line density of 3×1013 cm−2. Based on the measured plasma parameters, radial profiles of ion-energy losses due to classical processes are evaluated and ECRH in the central region is confirmed to reduce the energy loss due to electron drag significantly in the core-plasma region.

Noncontact minority carrier lifetime measurement of Si and SiGe epilayers prepared by ultrahigh vacuum electron cyclotron resonance chemical vapor deposition

A 70 MHz inductively coupled rf bridge probe is used to measure the minority carrier lifetime of the Si and SiGe epilayers grown by ultrahigh vacuum electron cyclotron resonance chemical vapor deposition (UHV-ECRCVD) at temperatures below 560 °C. Proper surface treatments before HF immersion are required for the accurate measurement of the bulk minority carrier lifetime. The effects of the process parameters such as the substrate dc bias, the distance of the ECR layer from the substrate, and the substrate temperature, including in situ surface cleaning, on the minority carrier lifetime of the Si and SiGe epilayers are examined by the rf bridge probe. It is confirmed that the rf bridge probe can monitor the epitaxial quality of low temperature Si and SiGe epilayers, making it an indispensable tool for the high quality device fabrication with Si and SiGe epitaxial layers grown by low temperature UHV-ECRCVD.

Effects of process parameters on low-temperature silicon homoepitaxy by ultrahigh-vacuum electron-cyclotron-resonance chemical-vapor deposition

The effects of the process parameters on the low-temperature Si homoepitaxial growth in an ultrahigh-vacuum electron-cyclotron-resonance chemical-vapor-deposition (UHV-ECRCVD) system are examined by reflection high-energy electron diffraction and transmission electron microscopy (TEM). The substrate dc bias during plasma deposition drastically changes the crystal structure from polycrystalline silicon at negative bias to single crystalline at positive bias. The defect production during plasma deposition is mainly caused by the energetic ions impinging on the Si substrate, and it can be effectively suppressed by the proper control of the process parameters in the direction of minimizing the ion energy. The positive substrate dc bias is a prerequisite for better crystallinity of low-temperature Si, but additionally the other process parameters such as microwave power, distance of the ECR layer from the substrate, SiH4 partial pressure, and total pressure should be definitely optimized to obtain dislocation-free Si epilayers. Dislocation-free Si epilayers are successfully grown at 560 °C at the positive dc bias greater than +10 V with the optimal control of the other process parameters. At temperatures below 470 °C, a high density of defects in the Si epilayers is observed by plan-view TEM, and the growth of the single-crystalline silicon is possible even without substrate heating but with a high density of defects. It is concluded that the substrate dc bias is a critical process parameter and the other process parameters do play a small but significant role as well in determining the crystallinity of the Si epilayers grown by UHV-ECRCVD.

Behavior of ECR microwave plasma formed in a mirror

Abstract Microwave plasma were produced and studied in a cylindrical chamber with a mirror and divergent magnetic fields. A circular TE 11 mode of 2.45 GHz was introduced into the plasma chamber via a dielectric vacuum window. It was found that microwave breakdowns occurred at sites where the static magnetic field was 875 G. There were two peaks of the plasma density and absorbed microwave power as a function of pressure in the range of 10 −3 −1 Pa. By moving the position of the ECR layer, it was found that the reflected microwave power had periodicity.