Abstract The effects of simultaneously generated ion and electron beams on silicon etching in an inductively coupled plasma system equipped with DC-biased grids were investigated. By adjusting the grid voltages, the energies of both ion and electron beams were precisely controlled, enabling charge-free etching through their simultaneous bombardment. Etching experiments on patterned silicon wafers with SiO ₂ masks revealed that concurrent ion–electron beam irradiation effectively suppresses plasma charging, thereby reducing bowing-like sidewall distortions in the etch profile. Furthermore, surface roughness measurements demonstrated that simultaneous ion and electron beam irradiation significantly decreased surface roughness compared to ion-beam-only etching. These findings indicate that the combined use of ion and electron beams mitigates plasma-induced damage and improves etching profile control, offering a promising method for high-aspect-ratio silicon etching.

Abstract The hydrogen dissociator is a critical component in a hydrogen maser for generating hydrogen atoms. The reliability of its ignition process governs the operational reliability of the maser, and its steady-state dissociation efficiency directly affects overall clock performance. However, plasma models for such sources have mostly focused on steady-state behavior, overlooking the complex transient evolution from breakdown to steady operation. Here we develop a zero-dimensional (0D) global model that follows the inductively coupled plasma (ICP) in a hydrogen maser dissociator from initial breakdown, through millisecond-scale relaxation, to the final steady state. The simulations show that the breakdown-field threshold depends strongly and non-monotonically on RF frequency. At low RF frequencies the breakdown field is highly sensitive to frequency changes, whereas at higher frequencies it varies much more gently, which helps explain the engineering preference for comparatively high operating frequencies and clarifies the trade-off between ignition efficiency and operational reliability. They further show that simply increasing input power is not an effective route to igher steady-state dissociation efficiency, because elevated temperature sharply enhances hydrogen atom recombination on the bulb wall and accelerate irreversible material aging. The model thus provides a new theoretical framework for dissociator design, shifting optimization from power escalation to refined strategies: reliable ignition via RF frequency selection and sustained efficiency via materials solutions that suppress wall recombination, offering practical guidance for operating-parameter optimization and long-term reliability improvement.

Abstract A novel time-dependent full-fluid 2D model of E × B discharges is presented. The model is based on the finite volume method with the addition of a threshold-based vacuum tracking algorithm. These tracking capabilities enable unprecedented fluid simulations of E × B plasma instabilities with near-vacuum regions. Two test cases are presented: the plasma layer (a semi-periodic domain resembling a hollow beam) and a square cross-section Penning-like discharge. In the two cases, an electron beam is injected into a background of neutrals, and ions are solely produced by impact ionisation. In both geometries, the simulations are performed from the limit of a pure electron plasma to a quasineutral plasma by increasing the background neutral density, showing a distinct transition from diocotron modes at low pressures to rotating spokes at high pressures. In line with theoretical and experimental results, the saturated regime of the diocotron mode results in the formation of electron vortices. The structure of the discharge in the different regimes is analysed, and the main spoke characteristics are assessed. In the square Penning configuration, the discharge transitions from vortices to the long-wavelength m = 1 rotating spoke. Finally, a variation of the square configuration is used to demonstrate the onset of the short-wavelength spiral structure observed in particle-in-cell simulations.

Abstract A microwave plasma source (MPS) is a device that generates high energy, high electron density plasma by ionizing gases with microwave energy (typically at 2.45 GHz), and is widely used in industry for its high energy density and ionization levels. However, their dynamic behaviors (driven by fluctuations in power, gas composition, and pressure) cause impedance mismatches, leading to nonlinear microwave reflection and reduced energy efficiency. To address this, we propose an MPS integrated with an electromagnetic black hole (EMBH). The EMBH operates based on a multilayer dielectric structure, which is designed to achieve omnidirectional absorption of electromagnetic waves. This unique structural design mitigates the impedance mismatch between microwaves and plasma, enabling automatic plasma adaptation and reducing energy reflection. Simulation results show that in a conventional MPS, the strong electric field standing wave node shifts away from the plasma upon excitation, whereas with EMBH incorporated, the electric field mode transitions from a standing wave to a traveling-standing wave, accompanied by a uniform distribution of the electric field around the plasma. Experiments under varying pressures (1-500 Pa), powers (200-800 W), and gases (argon, argon-hydrogen mixtures) confirm microwave energy conversion efficiency >85%. Notably, the system eliminates external tuning devices, enabling operational automation, reducing structural complexity, and improving reliability in dynamic conditions. These advantages facilitate compact design and efficient energy efficiency, demonstrating its broad industrial potential in materials processing, waste treatment, and beyond.

Abstract The effect of electrode quenching on improving CO2 decomposition in a coaxial DBD reactor is studied by combining experimental and numerical simulation methods. The experimental results reveal that rapid cooling of the reactor significantly improves CO2 conversion from 13.5% to the maximum 47.9%, and energy efficiency from 0.96% to 3.36% by activating internal electrode cooling water of 5°C after reaching thermal equilibrium. As the water cooling time further increases, the conversion and energy efficiency gradually decrease. Electrical characteristics and emission spectroscopy measurements show that electrode quenching reduces the gas temperature from 560 K to 440 K, and the burning voltage and reduced electric field first increase and then drop, which change the chemical kinetic processes. Zero-dimensional kinetic simulations reveal that electron direct impact dissociation is the dominant path for the maximum conversion close to 50%, since the reduced electric field strength increases by 18.7% at the initial rapid cooling stage. With the increase in water cooling duration, although the continuous decrease in gas temperature and the corresponding decline in three-body recombination rate CO + O + M → CO2 + M, the drop of reduced electric field strength leads to a decrease in the electron impact dissociation rate, reducing the CO2 conversion. Overall the decreased three-body recombination rate plays a small role during the cooling process, only leading to a slightly 20% higher conversion than the case without cooling at the final steady stage. These results highlight the importance of electrode quenching in improving CO2 conversion and provide insights for optimizing the DBD reactor.

Abstract This work investigates the O 3 generated in a planar atmospheric pressure dielectric barrier discharge (APDBD) reactor using Ar/0.5% O 2 as the working mixture with a microsecond pulsed power to analyze detailed species and reaction variations in the discharge zone. The numerical model integrates both the 1D plasma fluid model (PFM) and 3D gas flow model (GFM) to predict the O 3 generation in conjunction with the analysis of dominant species and mechanisms. The discharge uniformity is confirmed by the ICCD photograph taken, which justifies the implementation of the 1D PFM. The experimental measurements validate the simulated current, power consumption, reactor surface temperature, and O 3 concentrations at different regions. The uniform APDBDs using Ar/O 2 mixtures provide a simplified platform for exploring reaction mechanisms in Ar discharges under atmospheric pressure in contrast to their typical filamentary features. The analysis shows that atomic oxygen is the dominant species of O 3 generation, while excited oxygen molecules (i.e. O 2 (a) and O 2 (b)) are the dominant species of O 3 consumption. The variations of dominant generation and consumption reaction distributions reveal detailed interactions among species presented, leading to the differences between those predicted by the 1D PFM with all reactions and the integration of 1D PFM and 3D GFM. The developed numerical model provides an appropriate process considering both discharge dynamics and species transport behavior to predict variations of reactive species generated in the discharge zone.

Abstract To enhance feature-scale etching and deposition control, precise temperature regulation of wafer electrodes is critical in semiconductor manufacturing. Thermal gradients between wafers and chamber walls induce non-uniform background gas distributions, significantly altering plasma characteristics. This study employs a coupled 1D3V DSMC-PIC/MCC numerical model to investigate the effects of different electrode temperatures on capacitively coupled plasmas (CCPs) under low-pressure conditions. Non-uniform gas distributions generate asymmetric discharge behavior and induce a DC self-bias that depends approximately linearly on the temperature difference between the powered and grounded electrode. Consequently, ion fluxes, ion energy distributions (IEDs), and electron heating mechanisms exhibit significant modulation-a phenomenon that varies substantially with pressure.

Abstract We present a 2D axisymmetric fluid model consistently coupled with a 0D Boltzmann-solver electron energy distribution function and a detailed Ar/O₂ chemistry set that includes electron-impact dissociation and excitation, energy-transfer (quenching) reactions of O( 1 D ), ion neutral exchange, and wall recombination. Across 3–100 mTorr and 10%–90% O₂ mixing ratio, we quantify the spatial distributions of O( 3 P ) and O( 1 D ) and the O( 1 D ) flux at the target boundary. The O( 1 D ) density and flux are primarily set by dissociative excitation of O 2 (e + O 2 ( X 3 Σ g − ) → e + O( 1 D ) + O( 3 P )), electron-impact excitation of atomic oxygen (e + O( 3 P ) → e + O( 1 D )), and energy transfer from O( 1 D ) to O 2 (O 2 + O( 1 D ) → O( 3 P ) + O 2 (b 1 Σ g + )). In addition, the O( 1 D ) density and flux are modulated by (i) pressure-driven transport limitation (reduced diffusion and shorter mean free path), (ii) composition-dependent shifts between net production and O 2 quenching, and (iii) enhanced thermal speed and thermal diffusion at low O 2 due to elevated gas temperature. Consequently, high pressure and high O 2 density confine production to the under-coil inductive skin region and suppress the target-boundary flux, whereas at low O 2 density the weaker quenching and higher gas temperature partially offset transport-limited attenuation. These results indicate that predictive control of the O( 1 D ) density and flux requires a joint treatment of transport, explicit energy-transfer quenching, and thermal-diffusion effects within a detailed reaction set.

Abstract Precise control of ion energy distribution functions (IEDFs) is crucial for selectivity as well as control over sputter rate and substrate damage in nanoscale plasma processes. In this work, a low frequency (100 kHz) tailored pulse-wave-shaped bias voltage waveform is applied to the substrate electrode of an inductively coupled plasma (ICP) and its effects on the IEDF, electron density, electron dynamics and the etch rates of silicon dioxide as well as amorphous silicon are investigated in a commercial 200 mm reactive ion etching reactor. While the tailored waveform substrate bias hardly affects the electron density above the substrate and the spatio-temporally resolved electron power absorption dynamics, it is found to affect the ion flux to the substrate at high ICP source powers. Monoenergetic IEDFs with a full width at half maximum below 10 eV are realized with mean ion energies ranging from 20 eV to 100 eV in both argon and SF 6 . Using a modified voltage allows generating two independently controllable peaks in the IEDF. The monoenergetic IEDFs are used to determine the Ar ion sputter threshold energies of amorphous silicon and silicon dioxide to be 23 eV and 37 eV, respectively. This enables selective etching of these two materials by Ar ion sputtering based on tailoring the IEDF to ensure that all incident ions are within this narrow ion energy selectivity window.

Abstract The divergence topology of the applied magnetic field plays a critical role in plasma control and electromagnetic acceleration. To elucidate its impact on electromagnetic acceleration, we employed a dual-stage magnetic coil system for an applied field magnetoplasmadynamic thruster, and four field configurations characterized by different divergence ratios. The spatial coupling between plasma parameters and the magnetic-topology parameter B r / B z was subsequently analyzed under these varied divergence conditions. The result show that, under an overly divergent magnetic field configurations, insufficient radial confinement leads to a lower radial electron pressure gradient p e r , resulting in a diminished diamagnetic force. Additionally, premature electron demagnetization is induced by the divergent field, and the decelerating character of the paramagnetic E × B force is directly compromises the overall electromagnetic acceleration performance. Under overly convergent magnetic field configurations, the values of B r / B z in regions where p e r peaks remain low, resulting in a pronounced suppression of the diamagnetic electromagnetic force density. Therefore, there is a magnetic field configuration with an optimal divergence rate to achieve the optimal electromagnetic acceleration effect. Under the conditions of discharge current of 150 A and magnetic field strength of 50 mT, changing the magnetic field configuration ( k = 3.222) can increase the thrust of the thruster by 46% and the discharge voltage by 23%. The performance improvement is attributed to enhanced coupling between plasma parameters and the magnetic field under this configuration, which optimizes the conversion of electron pressure gradient and radial electric field into axial Lorentz force.