On phenomena in ionized gases
Statement of the problem
Download 9.74 Mb. Pdf ko'rish
|
- Bu sahifa navigatsiya:
- Challenges in the kinetic modelling of electrons and ions in gaseous and liquid matter
- 2. Self-consistent electron-biomolecule cross- section sets
- Advances and Challenges in Fluid Flow Models of Low-Temperature Plasmas Flows
- Electron temperature of thruster plume plasma in far field
1. Statement of the problem Modeling of plasma-surface interactions is a mul- ti-scale problem. At the lowest level lies the detailed atomic-scale description of the elementary acts of adsorption, desorption, diffusion, and reactions on the surface. Despite the huge progress in recent years on the computation of potential energy surfac- es from density functional theory and their use in classical molecular dynamics (MD), there is still a substantial discrepancy in the time and length scales of what is practically feasible in terms of calculation time at the atomic level and the time and length scales involved in a real system [1]. Another chal- lenge to MD calculations is to accurately account for excited states, electromagnetic fields, charged parti- cles and photons. At the next, mesoscopic, level, plasma-surface in- teractions can be modelled using stochastic kinetic Monte Carlo (KMC) algorithms [2]. KMC algo- rithms do not solve explicitly the master equation for a given system, but instead numerically simulate the underlying Markov process. Efficient algorithms describing NO and O 2 recombination in silica were recently presented [2], opening the door for a signif- icant development of this approach. However, these models lack a truly predictive power, as they need as input the energy barriers for each elementary step and other physical parameters. The KMC approach can be further coarse-grained to derive models adopting a deterministic descrip- tion (DD), where surface kinetics is formulated in terms of fractional coverages of different types of adsorption sites, simulated by a system of reaction- rate differential equations. The main advantage of this mesoscopic approach is its simplicity and com- putational efficiency, which allows the straightfor- ward coupling to gas phase chemistry in reactor- scale simulations and in computational fluid dynam- ics. However, compared to KMC, it does not ac- count for spatial correlations and cannot handle easi- ly probabilities that depend on the local configura- tion of the system, characterize fluctuations and relies on additional assumptions regarding the treat- ment of physisorbed species [2]. The incorporation of a description of surface modification under plas- ma exposure is another critical step for further de- velopment of both KMC and DD models. To bridge the gap between the sophistication of MD and the effectiveness of the DD remains per- haps the biggest challenge of all. In this context, KMC simulations play a central role: on the one hand, they can incorporate the information coming from ab initio simulations regarding binding ener- gies, energy barriers and dynamic surface modifica- tions; on the other hand, they can be used to validate and benchmark DD models and the underlying ap- proximations [2]. A combination of KMC and ab- initio calculations was already used for predictive modeling of real catalytic systems [3], but the appli- cation of this combined approach with generality remains another important challenge. A description of KMC methods and examples of application will be given at the conference. Acknowledgments: VG was partially supported by the Portuguese FCT, Projects UID/FIS/50010/ 2013 and PTDC/FIS-PLA/1420/2014 (PREMiERE)
[1] E.C. Neyts, Plasma Chem. Plasma Process.
Process. Polym. 14 (2017) 1600145. [2] V. Guerra and D. Marinov, Plasma Sources Sci. Technol. 25 (2016) 045001. D. Marinov, C. Teixeira and V. Guerra, Plasma Process. Polym. 14 (2017) 1600175. [3] M. Stamatakis, J. Phys.: Condens. Matter 27 (2015) 013001.
Topic n. 5 48 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Challenges in the kinetic modelling of electrons and ions in gaseous and liquid matter
R. D. White P 1 P , D. Cocks P 1,2 P , G. Boyle P 1
, M. Casey P 1 P , N. Garland P 1
P 1 P , J. de Urquijo P 3
, U M. J. Brunger P 2 P , R. P. McEachran P 4 P , S. J. Buckman P 4
, S. Dujko 5 , Z. Lj. Petrovic P 5 P
P 1 P
P
P
3 P
P
P
P
Modelling of electron and ion induced processes in plasma medicine and radiation damage is reliant on accurate self-consistent sets of cross-sections for electrons in tissue. These cross-sections (and associated transport theory) must accurately account not only the charged particle- biomolecule interactions but also for the soft-condensed nature of tissue. In this presentation, we report on recent swarm experiments for electrons in gaseous water and tetrahydrofuran using the pulsed-Townsend experiment, and the associated development of self-consistent cross-section sets that arise from them. We also report on the necessary modifications to the transport theory and gas-phase cross-sections required to accurately treat electron transport in liquids. The accuracy of the ab-initio theory is highlighted through comparison of theory and experiment for electrons in liquid Ar/Xe.
Accurate modelling of electron and ion transport in plasmas, plasma-liquid and plasma-tissue interactions is dependent on (i) the existence of accurate and complete sets of cross-sections, (ii) an accurate treatment of electron/ion transport in these phases, and (iii) accurate description of other processes e.g. localization (trapping), bubbles, etc. Modelling of electron/ions transport in gases, liquids and soft-condensed matter is considered through appropriate generalisations of Boltzmann’s equation to account for spatial-temporal correlations present in liquids including self-trapping of electrons into bubble states, and combined localised- delocalised nature of transport. Unified solutions of Boltzmann’s equation for electrons and ions are made within a multi-term framework, avoiding the well-known restrictions associated with the ‘two- term’ approximation.
The accuracy and completeness of electron- biomolecule cross-section sets can be assessed by comparison of calculated transport coefficients with those measured using a pulsed-Townsend swarm experiment of de Urquijo and co-workers. In this presentation we will present results from our recent studies of electrons in water, as the natural surrogate for human tissue. In addition, while DNA is currently not convenient to study, tetrahydrofuran (THF – C 4 H 8 O) has been investigated as a close analogue for low-energy electron interactions with 2-deoxyribose, a sugar that links phosphate groups in the DNA backbone.
As detailed above the treatment of electron transport in liquids involves distinctly more complicated physical processes than in the gas and crystalline phases. The randomness assumption inbuilt in the treatment of gases is no longer present, and neither is the long-range order generally present in crystalline materials. Rather in liquids there exists some short range order, where the scattering centres are spatially and temporally correlated. The impact of the screening of the electron interaction potential within the liquid is treated using an ab-initio solution of the Dirac-Fock equation, with a fully non-local treatment of exchange and accurate multipole polarisability in the electron-atom potential. We should emphasize that there are no adjustable parameters in the calculation [1]. In the presentation we will highlight our results for the transport of electrons in liquid argon and liquid xenon. Furthermore, we will highlight our preliminary results for electron capture into bubble states for atomic liquids [2].
[1] G J Boyle, R P McEachran, D Cocks, and R D White. J. Chem. Phys. , 142:154507, 2015 [2] D Cocks and R D White. arXiv:1602.07834v1
Topic number 49 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal 5 Challenges in PIC Modeling: Electromagnetic Description and Resonance Phenomena Thomas Mussenbrock Brandenburg University of Technology, D-03046 Cottbus, Germany This contribution provides an overview over applications of low-temperature plasmas for which new aspects of PIC modeling have to be taken into account. These new aspects come with a number of numerical as well as conceptual challenges, three of which are electromagnetic effects, resonance resonance effects, and plasma chemistry at high pressure. Boltzmann’s equation is certainly the most fundamental description of low-temperature plas- mas. It is in fact imperative to take it seriously when particle systems are not in thermal equilibri- um. [1] This holds in particular for low-pressure plasmas. It has been also recently shown that ki- netic effects even occur in atmospheric pressure plasmas. Under certain conditions electrons show strong non-Maxwellian behavior. [2] The energe- tic behavior of electrons is of course directly rela- ted to the ongoing plasma chemistry which makes low-temperature plasmas so useful. [3] Whenever accurate information about the energy distribution of particles in a non-equilibrium plasma is needed, e.g., to calculate the fundamental transport proper- ties of particles or the rates for elementary collision processes, a kinetic approach is mandatory. A number of different kinetic approaches for directly solving Boltzmann’s equation have be- en developed and used. One of these methods is particle-in-cell (PIC) coupled to Monte-Carlo col- lisions. [4 - 7] In this method super-particles in a Lagrangian frame – each of which represents mil- lions of real physical particles – are followed in continuous phase space whereas particle densities and current as velocity moments of the distributi- on functions are calculated on Eulerian grid points. The basic PIC method itself is intuitive and quite simple to implement. It consists of just four fun- damental procedures: i) integration of the New- ton’s equations of motion for the particles, ii) as- signment of charges and currents to the numerical field grid, iii) calculation of the fields on the grid points, and iv) interpolation of the fields from the grid to the particle positions. This straightforward and conceptually simple approach is probably one reason for its popularity, particularly in the low- temperature plasma simulation community. Although the PIC approach is more than 60 years old and quite straightforward – as briefly des- cribed above –, its development has not been com- pleted. With the today’s amazing applications of plasmas, new challenges in the context of numeri- cal plasma simulation using PIC pop up. This con- tribution is intended provides an overview over a limited number of applications of low-temperature plasmas for which new aspects of PIC modeling have to be taken into account. These new aspects come with numerical as well as conceptual chal- lenges, three of which are electromagnetic effects, resonance effects, and plasma chemistry at atmos- pheric pressure [8 - 10]. References [1] J.J. Duderstadt and W.R. Martin, Transport Theory (Wiley, 1979) [2] D. Eremin, T. Hemke and T. Mussenbrock, Plasma Sources Sci. Technol. 24, 044004 (2015) [3] A. Fridman, Plasma Chemistry (Cambridge University Press, 2002) [4] M.M. Turner, Phys. Plasmas 13, 033506 (2006) [5] J.M. Dawson, Rev. Mod. Phys. 55 403 (1983) [6] C.K. Birdsall and A.B. Langdon, Plasma Physics via Computer Simulation (McGraw-Hill, 1985)
[7] R.W. Hockney and J.W. Eastwood, Computer Simulation using Particles (Hilger, 1988) [8] D. Eremin, T. Hemke, R.P. Brinkmann, and Thomas Mussenbrock, J. Phys. D: Appl. Phys. 46, 084017 (2013) [9] S. Wilczek, J. Trieschmann, D. Eremin, R.P. Brinkmann, J. Schulze, E. Schuengel, A. Derzsi, I. Korolov, P. Hartmann, Z. Donkó, and T. Mussen- brock, Phys. Plasmas 23, 063514 (2016) [10] D. Eremin, T. Hemke, and T. Mussenbrock, Plasma Sources Sci. Technol. 25, 015009 (2016) 50
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
J.P. Trelles UP 1 P
P 1 P
Fluid flow models are essential tools for the analysis of low-temperature plasma (LTP) systems, especially in industrial application contexts. These models rely on the continuum approximation to describe the diverse range of chemical and thermodynamic nonequilibrium conditions inherent in LTPs. Numerical solutions face compound challenges found in other fields, such as nonlinearity of equation coefficients, resolution of large property gradients, instabilities and turbulence. These challenges are addressed by advanced methods designed for multiphysics and multiscale problems. Representative results of the use nonequilibrium fluid flow models for industrially relevant LTP systems are presented, which depict the challenges faced and approaches for their solution.
Low-temperature plasmas (LTPs) are at the core of diverse applications, such as materials processing, chemical synthesis, and medicine. The wide range of particle densities and energies in LTPs makes fluid models especially appealing for their description. The interaction of LTP with processing media, such as a gas stream or solid surface, leads to a chemical and thermodynamic nonequilibrium conditions. Such interactions also present complex coupling among fluid dynamics, heat transfer, chemical kinetics, and electromagnetic phenomena.
Numerical solutions of LTP fluid models face severe challenges found in other fields, including: - Resolution of large solution field gradients, due to, e.g. boundary layers, sheaths, or shocks, which are often conducive to instabilities and turbulence. - Handling of highly nonlinear equation coefficients and source terms result of constitutive relations, chemical reactions, Joule heating, etc., which produce numerical stiffness and limit convergence. - Consistent coupling of multiphysics model requirements, from the fulfilment of the solenoidal constraint of magnetic fields to the coupling between pressure and velocity fields in low-speed flows.
The above challenges are addressed by advanced numerical methods for multiphysics and multiscale problems. Stabilized and Variational Multiscale (VMS) methods exemplify the state-of-the-art among such methods, as evidenced by their use in commercial multiphysics software (e.g. Comsol) and their use in diverse fields, including plasma flows. Fig. 1 shows representative results of their use for the simulation of an arc torch, the core component in plasma spray; and the free-burning arc, a canonical model for the analysis of electric welding [1, 2].
Fig. 1: Fluid flow modelling of: (a) an arc plasma torch showing nonequilibrium between heavy-species (T h ) and
electron (T e ) temperatures, and (b) a free-burning arc, depicting the emergence of self-organized anode spots.
[1] J.P. Trelles, S.M. ModirKhazeni, Comput. Methods Appl. Mech. Engrg (2014) 282, 87-131. [2] J. P. Trelles, Plasma Processes and Polymers (2016) 14 (1-2), 1600092. 5 51
52 Oral Contributions 53
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal, 17
Electron temperature of thruster plume plasma in far field Boris Vayner Ohio Aerospace Institute, Cleveland, Ohio 44142, USA Voluminous arrays of data were obtained experimentally for various types of plasma thrusters operated in diverse chambers, and common conclusions were accepted: electron temperature decreased with distance from thruster orifice and increased with the background pressure decreasing. Plume plasma electron temperature is a very important parameter for evaluating the interactions between spacecraft elements and thruster plume. All measurements performed in vacuum chambers indicated rather low electron temperatures (0.5-2 eV) in the far field while computer simulations and measurements in space (one only) pointed to significantly higher temperatures (3-6 eV). The physical mechanisms of electron cooling in far field were not understood because of seemly collisionless electron gas in a vessel. It is shown in current paper that electron cooling in plasma chamber is caused by creation of potential barrier near walls, and this barrier originates from self-organization of electrically neutral plasma.
1.Introduction In order to perform tests in vacuum vessels one needs to know the parameters of the plume plasma in very far field. There are two complimentary approaches to the search for a solution to this problem: 1) performing extensive computer simulations; 2) measuring plasma parameters in ground chambers. These two approaches are mutually intertwined, but the results are frequently contradictive. Generally speaking, plume plasma parameters in a chamber and space are different: backpressure of neutral gas and vessel’s walls influence on plasma density, plasma
potential, and
electron temperature. The quantitative characteristics of these differences for any thruster depend on chamber dimensions and pumping speed [1]. The comprehensive study of all these factors was performed earlier. Background pressure (Xenon) varied from 3.5 µTorr to 73 µTorr. The electron temperature variations at the distance of 1 m from thruster exit plane (at the angle of 50 deg from axis) were determined within the range of 1-2 eV for floating thruster and 0.9-1.3 eV for thruster grounded. The electron temperature increased with pressure decreasing, and measurements error was estimated at 20%. In order to establish adequate test conditions the influence of a test arrangement on plume plasma parameters was analyzed and some criteria for appropriate ground test conditions were presented. 2. Ground experiments The effect of backpressure was studied for P5 Hall thruster in a quite large chamber with a diameter of 6 m and length of 9 m. Two sets of measurements were
performed at
xenon background pressures of 3.6 µTorr and 11 µTorr. Probes were positioned at the distance of 1 m from exit plane, which was equal to seven thruster diameters approximately (D
=148
mm). Certainly, ion current density was decreased about two times with increased pressure, and electron number density
demonstrated dependence on pressure with factors of 2-4. Electron temperature varied within the range of
=1.2-1.6 eV, and no correlations were established between electron temperature and neutral gas pressure. Plasma properties of Electron Cyclotron Resonance (ECR) thruster in the near field (2 cm from exit plane) were investigated. Electron temperature decreased with increasing flow rate: T e =2.5-3 eV at sccm m 20 , and T e =1.3-2.3 eV at
. 36 sccm m These results were obtained in fairly large chamber (D=2.2 m, L=7.9 m), and they confirmed that low electron temperatures were caused by processes inside the thruster but not the influence of background gas pressure. Plasma plume properties of the cluster of four BHT-200 Hall thrusters were measured at the distances comparable with assembly dimensions. Somewhat higher electron temperatures were recorded in far field for 1.5 kW Hall thrusters (PPS-100ML and PPI). 3. References
Download 9.74 Mb. Do'stlaringiz bilan baham: |
ma'muriyatiga murojaat qiling