On phenomena in ionized gases
Download 9.74 Mb. Pdf ko'rish
|
- Bu sahifa navigatsiya:
- 4. References
- Spatial and temporal analysis of acetone decomposition and subsequent OH formation in nanosecond diffuse discharge
- Figure 1.
- The influence of strong magnetic field on the plasma transport*
- 2. Resulting magnetized Fokker-Planck equation
- 3. Study of plasma transport essential processes
- Numerical modelling of stable glow corona discharges by means of stationary solvers of COMSOL Multiphysics
- 3. Acknowledgements
- -probe modeling for diagnostics of Plasma Propulsion Thruster
- 1. Introduction theory
- 2. Results conclusions
- Characterization of a ferro-electric packed bed plasma reactor
3. Results: Different microstructures (array of spots, lines and a group of letters) have been created on 50 nm Cu thin film using 18 – 30 keV Ar, Kr and Ne ion beams. For Ar ion beams, sputtering yield and milling rate are calculated at normal incident and found to be ~ 8.8 atoms/ion and ~ 0.65 µm 3 s
nA -1
respectively which are higher than Ga + focused ion beams (30 keV/1pA) for which calculated values are ~ 1.27 atoms/ion and ~ 0.09 µm 3 s
nA -1 . Next for further reduction of beam size, micro-glass capillary will be employed after PLE through which ion beam will pass, which will provide self-focusing of the beam without reducing the beam current [ 2 ,
]. For this first ion beam current will be measured by varying the extraction voltage (V
) using only capillary after PLE and then spot size of beam coming out from capillary will be measured. Microstructures will
be created
employing a capillary in the present ion beam system.
4. References [1] S. Bhattacharjee and S. Paul Jpn. J. Appl. Phys. 54, 01AA06 (2015). [2] S. Paul, A. Jayakiran and S. Bhattacharjee Appl. Phys. Lett. 101, 223508 (2012). [3] S. Paul and S. Bhattacharjee J. Phys. D: Appl. Phys. 48, 025204 (2015). 18
319 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Spatial and temporal analysis of acetone decomposition and subsequent OH formation in nanosecond diffuse discharge
K. Ouaras, L. Magne, P. Tardiveau, A. Brisset, S. Pasquier, P. Jeanney, B. Bournonville Laboratoire de Physique des Gaz et des Plasmas, CNRS, Paris-Saclay Université, 91400, Orsay, France
Planar laser induced fluorescence is employed to determine both the acetone (C 3 H 6 O) and the OH radical distribution during the post-discharge of a high voltage (85 kV), pulsed (10 Hz), nanosecond (10 ns), atmospheric pressure, centimeter gap (1.8 cm), diffuse pin-to-plane discharge operating in air containing different acetone concentrations (2000, 5000 and 10000 ppm). We determine both the spatial (in the whole interelectrode gap) and temporal (in post-discharge (t pd )) acetone Destruction and Removal Efficiency fraction DRE and the OH density [OH]. We emphasize both the largest acetone decomposition (~60 %) and the maximal OH density (5.10 16 cm
) in the pin region for the highest acetone concentration value (10000 ppm).
We investigated the spatial and temporal behavior of a pollutant, the acetone using Planar Induced Laser Fluorescence PLIF technic. Quantitative values concerning the destruction of the acetone and its conversion into OH are given. These spatial measurements are undertaken in a novel pin-to-plane discharge operated at very high voltage and short pulses. This study is intended to provide useful information about chemical kinetic in atmospheric plasma processes dedicated to environmental remediation.
The pin-to-plane discharge set-up [1] consists of a pin electrode powered by a nanosecond (10 ns), pulsed (10 Hz) high voltage (85 kV) power supply and a grounded plane electrode which are separated by 18 mm and mounted on a cylindrical stainless steel chamber equipped with optical windows in order to achieve PLIF measurement. The acetone is mixed at 2000, 5000 and 10000 ppm with dry air thanks to a bubbler system and the total gas flow rate is set to 1 L/min. PLIF experiments were performed in temporal post-discharge. The PLIF technic and the absolute calibration for OH density determination have been largely described in the literature [2] and will be not detailed here. Concerning the acetone processing, as the acetone is already introduced in the gas mixture, it consists of taking the LIF image without and with plasma discharge. The subtraction of these two images divided by the LIF image without plasma gives directly the quantitative DRE of acetone,
(1) where, [C 3 H
O] 0 and [C 3 H 6 O] are the initial and the final concentration of acetone, respectively.
The Figure 1 gives an overview of the spatial distribution profile in the interelectrode gap for air plus 10000 ppm of acetone mixture at 85 kV of (a) the acetone DRE and (b) the OH radical density.
Figure 1. Spatial distribution of (a) DRE of acetone at t pd = 200 µs and (b) OH density at t pd = 0.5 µs. 3. References [1] P Tardiveau et al 2016 Plasma Sources Sci. Technol. 25 054005
[2]
T Verreycken et al 2013 J. Phys. D: Appl. Phys. 46 464004
320 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
The influence of strong magnetic field on the plasma transport*
Chao Dong, Wenlu Zhang, and Ding Li †
P Institute of Physics, Chinese Academy of Science, Beijing 100190, China P
The full magnetized Fokker-Planck equation is derived through the transform method. The Fokker-Planck coefficients including the magnetic field are calculated by using the binary collision model. The magnetized Landau collision term is obtained. The influence of magnetic field on temperature relaxation has been extensively studied. It is shown that the strong magnetic field may greatly affect the electron anisotropic temperature relaxation, and electron-ion temperature relaxation in the tokamak edge plasma.
In many celestial and terrestrial environments, the particles’ gyro-radii are smaller than the Debye length. For example, in tokamak, the ratio of the thermal gyro-radius to the Debye length for electron could be much smaller than one for the plasma. The magnetic field affects the collision dynamics and associated transport phenomena such as velocity slowing down, temperature relaxation, diffusion, thermal transport etc. It was found that the cross field heat transport can occur even without mass transport when the magnetic field is very strong. [1]
The Fokker-Planck equation in the presence of a uniform magnetic field is derived through the transform method as follows:
where the Fokker-Planck coefficients 〈 ∆ 〉 and 〈 ∆ ∆ 〉 are calculated based within the binary collision model and the magnetized Landau equation is obtained:
The above kinetic equation is shown to be identical to the result obtained from the BBGKY approach when the collective effects are neglected and satisfy the conservation of particles, momentum, and energy. [2]
The strong magnetic field may greatly affect the transport essential processes in the plasma. It is shown that the electron-electron (e-e) and ion-ion temperature relaxation rates first increase and then decrease as the magnetic field grows, and the doubly logarithmic term contained in the electron-ion (e-i) temperature relaxation rate. [3] It is found that when the electron thermal gyro-radius is smaller than the Debye length, Debye length is replaced by the electron thermal gyro-radius in the Coulomb logarithm in the electron anisotropic temperature relaxation rate due to e-e collisions and e-i collisions. The e-i temperature relaxation rate contains a doubly logarithmic term arising from the exchange between the electron parallel and the ion perpendicular kinetic energies: [4]
For 10 19 3 , 3.5T ,
/ 3672 ,
2 , and lnΛ 15, we have lnΛ /lnΛ 1.37.
The other transport processes such as the velocity slowing down, diffusion, thermal conductivity and so on are being studied.
[1] M. Psimopoulos, D. Li, Royal Soc. Lond. A 437, 55 - 65 (1992). [2] Chao Dong, Wenlu Zhang, and Ding Li, Phys. Plasmas 23 (8), 082105, 2016. [3] Dong, Chao; Ren, Haijun; Cai, Huishan; and Li, Ding, Phys. Plasmas, 20 (10), 102518, 2013. [4] Chao Dong, Haijun Ren, Huishan Cai, and Ding Li, Phys. Plasmas, 20 (3), 032512, 2013.
† Email:dli@iphy.ac.cn * Supported by National Special Research Program of China For ITER and National Natural Science Foundation of China. 2) 321
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Numerical modelling of stable glow corona discharges by means of stationary solvers of COMSOL Multiphysics
P. G. C. Almeida P , N. G. C. Ferreira P , and M. S. Benilov P
P Departamento de Física, FCE, Universidade da Madeira, Largo do Município, 9000 Funchal, Portugal Instituto de Plasmas e Fusão Nuclear, Instituto Superior Técnico, Universidade de Lisboa, Lisboa, Portugal P
The use of stationary solvers for numerical simulations of DC gas discharges carries a number of advantages. This work describes modelling of stable glow corona discharges by means of stationary solvers of COMSOL Multiphysics. As an example, results are shown of calculation of a positive corona in a point-to-plane configuration.
The physics of glow (stationary) corona discharges has been understood reasonably well and a number of useful theoretical results, including analytical ones, have been obtained under various approximations. It is desirable to have also a fast and robust method of numerical modelling, which could be applied to a wide range of conditions. A standard approach relies on time-dependent solvers; e.g., [1,2]. Advantages offered by stationary solvers in simulations of DC discharges are demonstrated in [3]. In
particular, stationary solvers allow
computation of discharge modes in the whole range of their existence, thus decoupling physical and numerical stability, and are not subject to the Courant–Friedrichs–Lewy condition and
the corresponding limitations on the mesh element size.
As far as COMSOL Multiphysics is concerned, models of DC non-thermal discharges where no insulators are present can be implemented by using the so-called general or coefficient form or by means of using the Transport of diluted species and Electrostatics modules. The only way to accurately implement boundary conditions on the insulator is by means of the Plasma module, which has appropriate internal variables. However, a straightforward application of the Plasma module does not allow working with stationary solvers. The latter can be overcome by building a replica of the Plasma module in the weak form formulation [3]. This approach allows one to introduce also other relevant modifications, in particular, to allow the user to set diffusion coefficients of the ions. However, one loses access to the internal variables of the Plasma module while using this approach. In this work, the use of stationary solvers with the Plasma module was made possible by, paradoxically, setting equation form as time- dependent and manually controlling which
dependent variables are solved for. The above- mentioned modifications were introduced in the Plasma module by editing weak expressions and contributions. As an example, inception voltages, U
of the ionization integral, K, computed for the point- to-plane discharge configuration with 1 cm gap [4], are given in Table 1. Also shown are data computed without photoionization, U i (γ) and
(γ) , and the value of ln(1+γ ⁻¹).
γ U i (kV)
K U i (γ) (kV)
(γ)
ln(1+γ ⁻¹) 0 12.76 9.58 - - - 10 -4 12.74
9.54 14.41
12.85 9.23
10 -3
12.59 9.25
13.29 10.58
6.91 10 -2 11.91
8.01 12.08
8.31 4.62
10 -1
10.72 6.03
10.76 6.09
2.40
One can see that K (γ) >ln(1+γ ⁻¹); in other words, the Townsend breakdown condition does not apply. As γ increases, a transition from corona to Townsend discharge occurs as the role of dominating secondary electron production mechanism passes from the secondary electron emission to photoionization.
The work was supported by FCT of Portugal through the project Pest-OE/UID/FIS/50010/2013.
[1] P. Dordizadeh et. al., Plasma Sources Sci.
[2] L. Liu and M. Becerra, J. Phys. D: Appl. Phys. 49 (2016) 225202; 50 (2017) 105204. [3] P. G. C. Almeida et. al., Plasma Process Polym, DOI: 10.1002/ppap.201600122 (2017). [4] A. A. Kulikovsky, Phys. Rev. E
7066.
Topic 10 322
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
??????×??????-probe modeling for diagnostics of Plasma Propulsion Thruster
Timofey Chernyshev P 1 P , U Dariya Krivoruchko UP 2 P , Alexander Skrylev P 2
P 1 P
P
P
Plasma propulsion thrusters (PPT) are actively used in space. However presence of multiply charged ions (MCI) at PPT plume adversely affects main thruster parameters: thrust, mass utilization and lifetime. One of the instruments to measure MCI population is ??????×??????-probe [1]. Analyzing probe spectrum we assess MCI concentration and its velocity (or energy) distribution function (IVDF/IEDF). This diagnostic has been used on different PPT [2]. However for sources with a wide spread of ions velocity (in particular HT), it is a hard to predict probe’s parameters needed to resolve peaks for particles with different charges. In this work we present a model of the probe that has been created to predict probe transfer functions and methods for peak separation.
The interpretation of experimental data made with help of integral methods introduces inaccuracy and does not allow us to recover original IDVF due to asymmetrical broadening of the spectrum that depends on main particle velocity and blending of current peaks related to different ion species. In the work [3] the Fredholm equation for energy spectrum was used for data analysis to recover initial parameters of plasma with needed accuracy. Authors used Gaussian fitting for the raw signal for peak resolution problem. In the present work introduced the methods to determine ??????×??????-probe 's parameters. For this reason we made a program module that allows to calculate needed parameters of the probe and to decode measured spectrum. Let's define initial ??????-charged ion velocity distribution in probe’s axis direction as ??????(??????) and filtration speed as ?????? = ??????/?????? where ?????? – electric field and
?????? – magnetic field. The probe cuts out a small part of the initial distribution function ??????(??????) → ??????(??????, ??????). We can declare the probe's transfer function as ??????(??????, ??????) = ??????(??????, ??????)/??????(??????). The probe resolution ?????? is determined by the speeds where ??????(??????, ??????) becomes zero. The ion current to the collector surface is defined from Fredholm integral equation as ?????? = ???????????? with kernel ?????? = ?????? ⋅ ????????????(??????, ??????). The kernel of this equation can be obtained from the probe model. Then IVDF can be reconstructed by solving inverse problem with help of regularization methods. 2. Results & conclusions The numerical and analytical probe models were created to predict transfer functions, probe resolution and IVDF. These models show that probe’s resolution (in velocity units) depend on ??????
2 /??????
3
multiplied by geometric constant. In other words common integral interpretation of IVDF is incorrect for high-speed particle fractions. The calculated ions
in fig.1. Modeled broad signal demonstrated in fig.2.
Fig.1. Transfer functions for single charged ions. Fig 2. Simulated ??????×??????-signal for defined IEDF. Also, the
probe modification was suggested to better peaks separation. The modeling results are in a good agreement with available experiments data. 3. References [1] Sang-Wook Kim. Experimental investiga- tions of plasma parameters and species dependent ion energy distribution in the plasma exhaust plume of a hall thruster. PhD thesis. University of Michigan, 2001. [2] Wensheng Huang et al. Farfield Plume Measurement and Analysis on the NASA-300M and NASA-300MS. Tech. rep. NASA, 2013. [3] Youbong Lim et al. Observation of a high- energy tail in ion energy distribution in the cylindrical Hall thruster plasma. Physics of Plasmas 21.10 (2014), p. 103502. Topic number 5 323
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Characterization of a ferro-electric packed bed plasma reactor
A.M. Montoro-Damas 2 , A. Gómez-Ramírez P 1, 2
P , V. Rico 2 , A. R. González-Elípe 1 , J. Cotrino 1, 2
P 1 P
Departamento de Física Atómica, Molecular y Nuclear, Universidad de Sevilla, Avda. Reina Mercedes, 42022 Sevilla, Spain. P
P
The influence of diverse experimental parameters on characteristics of ferro-electric packed-bed plasma reactor used for hydrogen generation was investigated experimentally. The plasma reactor consisted in two parallel circular metal electrodes (Aluminum), and spherical shaped ferro-electric pellets packed in the discharge area. Barium Titanate (BaTiO 3 ) and (Lead Zirconate Titanate, Pb[Zr x Ti 1-x ]O 3 (0 ≤ x ≤ 1)) PZT was used as ferro- electric materials. Sinusoidal high voltage up to a maximum of 2.5 kV was applied to the upper electrode with a frequency range between 50Hz and 10kHz. The determination of electrical parameters (such as instantaneous power, transferred charge, breakdown voltage, electron density and capacitance properties) was carried out in different reactor configurations. Download 9.74 Mb. Do'stlaringiz bilan baham: |
ma'muriyatiga murojaat qiling