On phenomena in ionized gases
Download 9.74 Mb. Pdf ko'rish
|
- Bu sahifa navigatsiya:
- Direct kinetic simulation of nonlinear plasma waves and Hall thruster discharge plasmas
- 1. Background and Motivation
- 3. References
- Rotating spoke instabilities in standard and wall-less Hall thrusters: Experiments and PIC simulations
- Dynamic of HiPIMS Plasmas
- Challenges in the modelling of reactive plasmas: limitations and opportunities in global modelling
- 2. Overcoming spatial and temporal averaging
- References
- Challenges in the modelling of plasma-surface interactions
3. References [1] S. Vigne et al., Opt. Mater. Express 5 (2015) 2404. [2] V.R. Morrison et al., Science 346 (2014) 445. [3] N. Emond et al., Appl. Phys. Lett. 107 (2015) 143507. [4] N. Emond et al., Appl. Surf. Sci. 379 (2016) 377.
40 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Direct kinetic simulation of nonlinear plasma waves and Hall thruster discharge plasmas
K. Hara 1
P 1 P
Plasma instabilities and oscillations play an important role in many plasma applications ranging from wave- and beam-plasma interactions to plasma propulsion. In order to investigate such nonlinear plasma phenomena, we have developed a direct kinetic (DK) simulation, in which the velocity distribution functions are directly obtained by solving kinetic equations such as the Vlasov equation in a discretized phase space, i.e., physical and velocity space. One advantage of a DK method is that the numerical noise inherent in particle methods is essentially eliminated. We employed the DK simulation to investigate plasma-wave interactions, including electron plasma waves, ion acoustic waves, and trapped particle instabilities, and the low-frequency ionization oscillations in the discharge plasmas of Hall effect thrusters.
Two main plasma simulation techniques that have been developed include fluid and kinetic models. Fluid models solve the conservation equations for macroscopic quantities, including the density, bulk velocity, and energy. On the other hand, the first- principles gas kinetic equations, such as the Vlasov and Boltzmann equations, are solved to obtain the distribution functions in kinetic models. One of the most popular is particle methods, in which computational “macroparticles” are used. However, the statistical noise inherent in the particle methods may alter the physical oscillation signals if the number of macroparticles is not sufficient. We have developed a grid-based kinetic simulation, called the direct kinetic (DK) method, which is an alternative to particle-based kinetic models such as particle-in-cell (PIC) method. As the kinetic equations are hyperbolic partial differential equation, we employ a finite-volume method using Monotonic Upwind Scheme for Conservation Laws (MUSCL) framework with the Arora-Roe limiter, which preserves conservation and positivity of the distribution functions. The DK method has been tested against plasma- sheath theory with and without secondary electron emission, Landau damping, and nonlinear plasma wave theories for Langmuir and ion acoustic waves. It is also benchmarked with a PIC simulation in the Hall thruster discharge plasma. [1] We developed a hybrid-kinetic simulation, in which ions are solved using a kinetic (DK or PIC) simulation while a fluid model is used for electrons, and showed that DK simulation is useful for ionization oscillations.
I will focus on the two nonlinear phenomena that we investigated using a DK method. First, we performed kinetic simulations of the trapped particle bunching instability in nonlinear plasma waves. [2] We have shown that the trapped particles in traveling potential wells experience a bunching instability and form a bunch in the phase space depending on the initial trapped particle distribution. The growth rates obtained from the numerical simulation are in good agreement with the theoretical predictions. Second, low-frequency (10-30 kHz) ionization oscillation, also called the breathing mode, in Hall effect thrusters is investigated. Although this phenomenon has been often observed in experiments and numerical simulations, the mechanism of excitation and damping of the breathing mode was not fully understood. We have employed a hybrid-DK method and showed a qualitative agreement of the discharge current oscillations between experiments and numerical simulations. [3] Furthermore, we developed a perturbation theory of ionization oscillations by accounting for the perturbation of electron energy, which was neglected in the conventional predator-prey type formulation. [4] From the numerical results and the perturbation theory, we have concluded that the electron heat transfer mechanism plays a significant role in the mode transition of the ionization oscillations.
[1] K. Hara, I. D. Boyd, and V. I. Kolobov, Phys. Plasmas 19, 113508 (2012) [2] K. Hara, T. Chapman, J. W. Banks, et al., Phys. Plasmas 22, 022104, (2015) [3] K. Hara, M. J. Sekerak, I. D. Boyd, and A. D. Gallimore, J. Appl. Phys. 115, 203304 (2014) [4] K. Hara, M. J. Sekerak, I. D. Boyd, and A. D. Gallimore, Phys. Plasmas 21, 122103, (2014) 4 41 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Rotating spoke instabilities in standard and wall-less Hall thrusters: Experiments and PIC simulations
S. Mazouffre, L. Grimaud, S. Tsikata, K. Mathyash, R. Schneider P
P
P
P
This work reports on the examination on rotating plasma inhomogeneities, also called rotating spokes, in the discharge of a low-power Hall thruster. Rotating structures have been observed by means of high-speed imaging and time-resolved LIF spectroscopy for two configurations of the thruster: the standard configuration and the wall-less configuration in which the plasma discharge is unbounded. Numerical simulations based on a 3D PIC model support the experiments.
Rotating plasma inhomogeneities are observed in various types of magnetized low pressure plasma discharges created in a crossed electric and magnetic field configuration. Over the past ten years, such structures, often termed “rotating spokes”, have been experimentally investigated e.g. in plasma devices like magnetrons and Hall thrusters. There is a great deal of interest in studying rotating plasma structures for mainly two reasons. Firstly, such instabilities seem to be a very general phenomenon in low-pressure plasma discharges of which the origin is not yet fully understood. Secondly, these large-scale low-frequency rotating instabilities certainly play a role in the transport of charged particle. As a consequence, they probably influence both the characteristics and the performances of plasma devices and plasma technologies like thin- film deposition and spacecraft propulsion.
In this contribution we experimentally examine the properties of low-frequency (a few kHz) rotating plasma instabilities in the discharge of the low power ISCT200 Hall thruster. The latter is a versatile 200 W-class Hall thruster using permanent magnets for generating the magnetic field instead of helical
magnetizing coils.
Two different configurations of the ISCT200 have been employed in this study. The standard (ST) one rests upon a magnetic barrier perpendicular to the cavity walls. The unconventional wall-less (WL) configuration allows to entirely shift the plasma discharge outside the cavity, then eliminating wall processes such as secondary electron emission and sputtering. The two versions, however, share many common features: the channel geometry is the same, walls are made of BN-SiO
2 ceramic, the magnetic field is produced by SmCo magnets, a porous compound serves as propellant gas injector and a heated cathode with a disk-shaped LaB 6 emitter provides the necessary electron current for maintaining the discharge and neutralizing the ion beam. In ST configuration, a ring anode is placed at the back of the channel whereas in WL configuration, a gridded anode with circular holes is placed at the cavity exit plane.
The main objective of this work is to characterize the physics and the dynamics of rotating plasma structures in the E×B discharge of a 200 W-class Hall thruster operating with xenon by means of two diagnostic techniques. High-speed camera imaging has been used to capture the rotating spoke motion and transformation for various thruster operating conditions. Image processing with sophisticated algorithms allows to determine the rotation velocity, direction and frequency, the plasma structure shape and sizes and the mode number. Time-resolved LIF spectroscopy in the near infrared has been employed to record the temporal evolution of the Xe + ion azimuthal velocity component during the rotation of a plasma inhomogeneity. Here, a novel photon counting approach has been developed to enable
measurements without
externally stabilizing the discharge. The following points will be presented and discussed: the determination of the properties of such structures (velocity, frequency, domain of existence) and correlation with longitudinal instabilities such
as breathing oscillations, the impact of the discharge voltage, propellant mass flow rate and cathode heater current on the properties of the rotating spokes, clarifications as to the origin of the rotating inhomogeneities, time evolution of the ion velocity distribution function in the course of a spoke rotation, and finally a critical comparison of the features of rotating structures in conventional and wall-less Hall thrusters. Experimental results are supported by outcomes of numerical simulations carried out with a 3D PIC model. Topic number 42
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Dynamic of HiPIMS Plasmas
A. von Keudell, C. Maszl, W. Breilmann, Julian Held, V. Schulz-von der Gathen, A. Hecimovic
P 1 P
ranging from optical emission spectroscopy, probe diagnostics to mass spectrometry. It shown that structure formation in these plasmas is driven by the Simon-Hoh instability leading to the appearance of rotating spokes along the racetrack of the magnetrons. The plasma parameters in these rotating ionization zones are measured using time resolved optical and mass spectrometry. It is shown that the energy distribution of the ions reaching the substrate are directly connected to the appearance of the spokes. The underlying mechanisms are discussed to explain the good performance of HiPIMS plasmas for material synthesis
High power impulse magnetron sputtering (HiPIMS) plasmas are characterized by a high degree of ionization and a very energetic metal growth flux leading to superior material properties. Power densities at the target of several kWcm -2 are realized by using short pulses of 10 to 200 µs and duty cycles of a few percent only. Many studies focus on unraveling the dynamic of a HiPIMS plasma. The intense sputter wind in a HiPIMS pulse causes gas rarefaction after a time span of 10...30µs after the onset of the plasma pulse. At target power densities above 1 kW cm -2 , localized ionization zones, so-called spokes, are observed which rotate along the plasma torus with a typical velocity of 10 km s -1 . It is assumed that the localized ionization zones correspond to regions of high electrical potential, and are, therefore, the source of an energetic group of ions of typically few tens of eV in the growth flux on the substrate. The spoke pattern depends on target material, plasma gas, power density and pressure. By adding a reactive gas such as oxygen or nitrogen to a HiPIMS plasma specific oxides and nitrides can be deposited on the substrate. The analysis of the plasma parameters of this dynamic plasma pulses is very demanding because two time scales need to be taken into account: the first is the pulsing of the discharge with a duty cycle of a percent of less, the second is the dynamic of the plasma evolution during each pulse itself, where the current is rising from zero to over 100 A for a 2” target and the spokes form dynamically. Therefore, an elaborate triggering scheme is developed, to trigger the optical diagnostic as well as the mass spectrometer to the presence of the ionization zone in the direct line of sight to the diagnostic. By using Stark broadening of hydrogen lines, the electron density inside the spokes can be determined to 10 14
-3 . By inserting probes into the magneton target, the modulation of the target current by the traveling ionization zones of 30% could be determined. The synchronized mass spectra show that the energetic ions are uniquely connected to the presence of the spokes. This can be explained by the occurrence of a Simon-Hoh instability which modulates the electrical potential in the plasma and thereby the energy of the ejected ions.
[1] W. Breilmann, C. Maszl, A. Hecimovic, A. von Keudell, J. Phys. D (accepted) (2017) [2] A. von Keudell, A. Hecimovic, C. Maszl, Contrib. Plasma Phys. 1, 9 (2016) [3] A Hecimovic, C Maszl, V Schulz-von der Gathen, M Böke and A von Keudell, Plasma Sources Science and Technology 25, 035001 (2016) [4] A. Hecimovic, V. Schulz-von der Gathen, M. Böke, A. von Keudell, Plasma Sources Science & Technology 24, 045005 (2015) [5] W Breilmann, A Eitrich, C Maszl, A Hecimovic, V Layes, J Benedikt, A. von Keudell, J. Phys. D 48, 295202 (2015)
43 44 Special Session Lectures 45
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Challenges in the modelling of reactive plasmas: limitations and opportunities in global modelling A. R. Gibson 1,2
1 York Plasma Institute, Department of Physics, University of York, Heslington, York, YO10 5DD, UK 2 LPP, CNRS, Ecole Polytechnique, UPMC Univ. Paris 06, Univ. Paris-Sud, Observatoire de Paris, Université Paris-Saclay, Sorbonne Universités, PSL Research University, 91128 Palaiseau, France
The use of global models in various forms is commonplace in the low temperature plasma community in large part because of their computational simplicity, which leads to short solution times and relative ease of interpretation. However, the penalties paid for a short solution time are often the loss of spatial and temporal information on the system of interest. In some applications, these variations could be considered negligible, but in others, they are crucial in defining the properties of the plasma. In this contribution, a perspective will be given on the limitations of global models, mainly from the point of view of temporal and spatial averaging. Specific examples of global models developed to circumvent these limitations will be presented along with a further perspective on the opportunities presented by such approaches for the field of global modelling, particularly with a view to improving comparisons with experimental measurements.
1. Introduction Global models are often the first step undertaken by researchers aiming to understand complex variations in plasma properties under the change of a given external parameter. The low degree of computational complexity involved in these models allows for studies of systems inclusive of complex gas mixtures and reaction mechanisms, which are significantly more difficult using higher dimensional models. The lack of spatial dimensions and analytical complexity additionally simplifies the analysis of the results of the model allowing complex phenomena to be identified even in models inclusive of large reaction mechanisms.
The spatially and temporally averaged assumptions inherent in the most basic global model approaches present clear limitations when applied to many
experimental systems, meaning that
comparisons between
global models
and experimental results are often difficult. However, several works have demonstrated that it is possible to extend the basic global model approach to more complicated experimental systems while
maintaining computational simplicity provided that the dominant physical properties of these systems are understood [1, 2, 3].
This contribution will discuss examples where the assumptions of spatial
and temporal homogeneity inherent in global models limit the understanding of important phenomena in certain physical systems and how these limitations can be overcome while maintaining computational simplicity [4].
A further perspective will be given as to the opportunities presented to the field of global modelling by such approaches. Particular emphasis will be given to how these approaches may improve comparisons between global
models and
experimental results for systems that cannot be reasonably viewed to be spatially or temporally homogenous.
Acknowledgements This work has been done within the LABEX Plas@Par project, and received financial state aid managed by the ‘Agence Nationale de la Recherche’, as
part of
the ‘Programme d’Investissements d’Avenir’ under the reference ANR-11-IDEX- 0004-02. Funding through UK EPSRC (EP/ K018388/1) and the York-Paris Low Temperature Plasma Collaborative Research Centre is also acknowledged.
[1] E. Kawamura, M. A. Lieberman, A. J. Lichtenberg, P. Chabert, C. Lazzaroni, Plasma Sources Sci. Technol., 23, (2014) 035014.
[2] K. Niemi, T. Gans, D. O’Connell, Plasma Sources Sci. Technol., 22, (2013) 032001. [3] E. Despiau-Pujo, P. Chabert, Plasma Sources Sci. Technol. 18 (2009) 045028
[4] A. Hurlbatt, A. R. Gibson, S. Schröter, J. Bredin, A. P. S. Foote, P. Grondein, D. O'Connell, T. Gans, Plasma Process Polym, 14 (2017)
Topic number 5 47 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Challenges in the modelling of plasma-surface interactions
Vasco Guerra 1 P and Daniil Marinov 2
1 Instituto de Plasmas e Fusão Nuclear, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, Portugal 2 imec, Kapeldreef 75, 3001 Leuven, Belgium
This contribution briefly addresses some current challenges in the modelling of plasma-surface in- teractions. Different types of models focus on different time and length scales and have specific challenges of their own. From the shorter to the longer scales, these formulations include atomic level simulations, kinetic Monte Carlo algorithms (KMC) and deterministic descriptions. The big- gest challenge, however, resides perhaps on how to include the detailed information of atomic scale simulations into reliable and effective models that can be easily incorporated in simulations of realistic systems. A key role can be played by the KMC methods, currently under expansion, as their intermediate degree sophistication places them in a perfect position to help bridging this gap.
Download 9.74 Mb. Do'stlaringiz bilan baham: |
ma'muriyatiga murojaat qiling