On phenomena in ionized gases
Download 9.74 Mb. Pdf ko'rish
|
|
2. Results Discharge voltage, discharge current and energy per pulse and the corresponding frames of the discharge are shown in Fig. 2. Three main behaviors of N 2
Step 1: the plasma is located on the sandwich throughout the negative pulse of current (0.5 mA) and voltage (800 V). Step 2: at 1 µs the cathodic inversion occurs. The radiative zone moves at a speed of ~ 80 km.s -1 , which is typical of the velocity of a streamer. Step 3: at the voltage decay, the overlapping jets are observed until disappearance, with a velocity of 100 m.s
-1 , which corresponds to the gas velocity.
Fig.2. Main steps of the N 2 plasma propagation in the low pressure side of the reactor, including electrical signals
Using a fitting procedure of the second positive system of N 2 at 337 nm [3], we estimate the rotational temperature to be 850 ± 20 K, and the vibrational temperature equal to 3210 ± 200 K under the same conditions.
[1] K.H. Schoenbach et al., Appl. Phys. Lett. (1996) 68, 13-16 [2] C. Lazzaroni et al., Eur Phys. J. D (2010) 60, 555-563 [3] K. Gazeli et al., J. Appl. Phys. (2015) 117, 093302
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Towards a fluid model for the streamer-to-leader transition in lightning channels
A. Malagón 1 , A. Luque 1
1 Instituto de Astrofísica de Andalucía (IAA), CSIC, Granada, Spain
One of the still unknown phenomena involved in electric discharges is the streamer-to- leader transition. Here, as a first approach to investigate this transition, we present our recent steps modelling heating and gas expansion produced in streamer discharges, coupled to electromagnetic and air plasma chemistry. So far, we have applied this model to the mechanism of air heating and pressure perturbations present in sprite discharges, without losing sight of our final aim, i.e., to disentangle the streamer-to- leader transition in lightning channels by testing whether attachment instability leads to heating of air to a temperature high enough develop space stems.
on Earth's atmosphere. However some of their features are still poorly understood. A sufficiently long electric discharge, such as a lightning channel, propagates along two phases. The first phase is known as “streamer phase” and consists in thin filaments of ionised air that advance due to a high electric field at their tip. The dominant process of ionisation is impact ionisation, involving electrons and the two major components in the air mass, which are nitrogen and oxygen. In the second phase called “leader phase”, the electric current of the streamers has increased the air temperature highly enough so the thermal energy of the molecules present in the air is comparable to the ionisation potential of nitrogen and oxygen. The underlying mechanism whereby the streamer-to-leader transition occurs is not precisely known. High-speed observations show that in negative discharges, comprising 90% of cloud-to-ground lightning, this transition is not smooth but rather mediated by the formation of a “space leader”, that is, an isolated hot segment within the streamer region. This space leader is connected to the main leader in a sudden jump and therefore one speaks of a “stepped leader”. However, the origin of the space leader is so far unknown. Here we present recent steps in the modelling of the streamer-to-leader transition, which requires coupling fluid mechanics, electromagnetism and air plasma chemistry. We discuss our work towards a model that solves Euler's equations (3 dimensions reduced to 2 by virtue of symmetry) coupled to electron drift using high-resolution finite volume methods for hyperbolic systems [1] implemented in the software package CLAWPACK. The drift of electrons is determined by a self-consistent electric field, which we obtain by solving Poisson's equation by means of off-the-shelf solvers. Our model also includes a selection of chemical reactions that have a relevant effect on the electron density in air, such as impact ionisation, attachment and detachment. Besides the streamer-to-leader transition, our model can also be applied to the mechanism of air heating and pressure perturbations caused by sprite discharges in the upper atmosphere. We present here preliminary simulation results on this topic. Our final aim is however to test the hypothesis that leader stepping results from an attachment instability that creates low-conductivity, high-field regions in a streamer corona, as recently discussed for sprites in [2]. With our detailed model for gas heating and expansion we will investigate whether the attachment instability leads to heating of air to a temperature high enough to develop space stems. A positive answer to this question would elucidate the physical mechanism of leader stepping.
[1] R.J. LeVeque. Finite Volume Methods for Hyperbolic Problems. Cambridge Texts in Applied Mathematics. Cambridge University Press, 2002.
and R.K. Haaland. Srpite beads and glows arising from the attachment instability in streamer channels. J. Geophys. Res. (Space Phys), 121, 2016.
7 292 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Sensitivity and uncertainty analysis of a kinetic model for CO 2
non-equilibrium plasmas
Marija Grofulović, Tiago Silva and Vasco Guerra Instituto de Plasmas e Fusão Nuclear, Instituto Superior Técnico, Universidade de Lisboa, Portugal
This work is dedicated to the systematic investigation of the influence that uncertain rate constants have on model predictions. The kinetic scheme examined here describes CO 2 plasmas with a very complex vibrational kinetics model. Uncertainty of the rate constants models arise from the experimental errors or the approximate theories. To rank the most influential rate constants we used the Morris method, that is a One-At-a-Time design for the sensitivity analysis (SA). The output parameters that were traced are the densities of some species, i.e. CO 2 (001) and CO 2 (010)
and the vibrational temperatures T 1,2
and T 3 . The results show that the dominant e-V reactions are also the ones with larger contribution to the uncertainty of the output.
1. Introduction Modelling low-temperature plasmas usually involves very complex chemistry models described by data that are measured or calculated with some uncertainty. In order to make the simulations of experimentally difficult or inaccessible conditions more credible, the influence of the input uncertainty on the output must be studied [1].
Figure 1 The result of SA, µ* [1] for the vibrational temperature T v3 a); and the simulation results using varied rate constants for e-V, V-V or V-T processes b). A good example of such systems is CO 2 plasma. Recent interest in CO 2 conversion has set the goals for fundamental experimental research and plasma modelling. One particularity of these plasmas is the extended vibrational kinetics, which can comprise ~9000 vibrational levels in a full state-to-state description. In the present work, the SA was performed on a kinetic scheme describing a low excitation regime in CO 2 , accounting for 72 individual vibrational levels and more than 1200 elementary processes. The operating conditions are well characterized experimentally, allowing the validation of the model presented in [2].
2. Results and discussion The screening procedure used in this work, proposed by Morris [3], gives both the quantitative and the qualitative information on the influence of the input uncertainty. Figure 1a) shows that the e-V reactions contributing to the uncertainty of the T v3
are the ones populating the asymmetric levels from the ground state. Different simulation results for rate constants varied within the ascribed uncertainty are shown in figure 1b). The present analysis evaluates the level of model reliability, reveals the possible sources of model failure and therefore recommends directions for future improvement.
Acknowledgments: This work was partially supported by FCT, Projects UID/FIS/50010/ 2013, PTDC/FIS-PLA/1420/2014 (PREMiERE) and grant PD/BD/105884/2014 (PD-F APPLAuSE).
References [1] M. M. Turner, Plasma Sources Sci. Technol. 24 (2015) 035027.
[2] T. Silva et al, contribution to ICPIG, (2017). [3] M. Morris, Technometrics, 33 (1991) 161–74 Topic number 5 293
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal Topic 6
Non-intrusive Method for Electron-Density determination in Low-pressure Microwave Plasma A.Kais 1
1 , L.Thérèse 1 and Ph.Guillot 1 1 Laboratoire Diagnostics des Plasmas Hors Equilibre, Université de Toulouse, INU Champollion, Albi, France The power supplied by the plasma at the surface of a glass substrate is measured and calculated. The total contribution of the heating mechanisms is calculated according to the theories commonly used in the literature, and measured by exploiting the temperature curve variation in the heating phase (plasma on). The cooling mechanisms are leads by the conduction with the gas and the sub- strate holder, their contribution is measured using the temperature variation during the cooling phase (plasma off). Assuming that our plasma obeys the hypothesis of the corona balance, the Modified Boltzmann Plot (MBP) method is used to determine the electron temperature T e . A correlation be- tween the power deposited by the plasma and the results of the MBP is established. This correlation indicate that it is possible to estimate the electron density (n e ) without using the Langmuir probe. 1. Introduction This work is dedicated to the determination of plasma electron-density without using the Lang- muir probe (intrusive method which can be unhandy in some cases). The method developed allows to calculate the electron density by combining two non-intrusive methods: the temperature measure- ment [1] and the modified Boltzmann plot [2]. 2. Experimental sut-up This study was carried-out in a pure argon gas at pressure range of 10 to 30 Pa. Plasma is generated by a coaxial microwave plasma source (Hi-Wave) switched-on by a 2.45 GHz Solid State Generator which the power can vary from 1 to 200 W. The parameters of the discharge are measured by a double Langmuir probe (Impedence Ltd). The temperature time variations are measured by a K- type thermocouple. Optical Emission Spectroscopy measurements are performed by an Avaspec 2048- 2-Avantes spectrometer with a resolution of 3 nm. 3. Experimental results In this work, we will shown that P th =P
, where P
th is the calculated power at the surface obtained by summing the contributions of electron, ions and electron-ion recombination [3], and P m is
ing the temperature curve variations. Thus, the fol- lowing relation between P m and n
e is established: n e
P m A s [√ k B T e 2πm e exp ( e 0 V sh k B T e ) (2k
B T e + E ion
) +0.3k
B T e √ k B T e M 1 2 k B T e | ln 2πm
e M + 1 | ] −1 (1) One can see in this equation that electron density depends on one unknown parameter (T e ). The de- termination of T e by the MBP method allows the calculation of the electron density. In equation 1, V sh represents the sheath potential taken equal to (k B T e /2e 0 )(ln(M/2πm e )
ion is the ionization energy of argon, M is the mass ion and A s is the total substrate surface. Figure 1: Electron density calculated from Eq.1 compared with Langmuir probe measurments: effect of the pressure (a) and the microwave power (b). 4. References [1] H. Kersten et al. Journal of Applied Physics 87 (2000) 3637 [2] F J Gordillo-V’azquez et al. Plasma Sources Sci. Technol.
15 (2006 ) 42 [3] Daniel Lundin et al. J. Phys. D: Appl. Phys. 42 (2009) 7
294 XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Relaxation of electronic excitation in nitrogen discharge plasma at high specific deposited energy
N.D. Lepikhin 1 , N.A. Popov 2 , S.M. Starikovskaia 1
1 Laboratoire de Physique de Plasmas (CNRS, Ecole Polytechnique, Sorbonne Universities, University of Pierre & Marie Curie-Paris 6, University Paris-Sud, Observatoire de Paris), France 2 Skobeltsyn Institute for Nuclear Physics, Moscow State University, Moscow, Russia
The energy relaxation from electronic degrees of freedom of molecular nitrogen excited by a capillary nanosecond discharge at high specific deposited energy and electric fields of 200-300 Td is investigated experimentally and numerically. The key role of pooling reaction between metastable N 2 (A 3 u ) states and quenching of N( 2 D) atoms by N 2 in the mechanism of fast gas heating at high specific deposited energy is demonstrated. The temperature dependence of pooling reaction rate constant was obtained based on the treatment of available experimental data.
The knowledge of the channels of energy relaxation from electronically excited states is extremely important for the applications. This study is dedicated to the experimental and numerical study of the relaxation of electronic excitation and fast gas heating in pure nitrogen, excited by capillary nanosecond discharge at electric fields of E/N = 200-300 Td and specific deposited energy
up to 1 eV/molecule.
The nanosecond discharge was initiated in the capillary tube 1.5 mm of internal diameter and 70 mm in length, P = 27 mbar. High-voltage pulses of U = + 9.3 kV, 30 ns FWHM were used to initiate the discharge. For each initial pulse from the generator, typically three pulses with progressively attenuated amplitude separated by 250 ns were observed. A detailed description of the experimental setup and diagnostic techniques are presented in [1].
3. Results and discussion The temporal evolution of gas temperature in nitrogen, excited by capillary nanosecond discharge, measured experimentally and compared with the one calculated numerically in 1-D axially symmetric model is presented in Fig. 1. The calculations were performed according to the model described in [1,2]. The observed fast increase of gas temperature in nitrogen (2200 K/μs, Fig. 1) demonstrates fast energy relaxation from electronic degrees of freedom. The pooling reaction between metastable N 2 (A 3 u ) states with 3.5 eV energy converted to translational energy [2] is found to be a dominate process responsible for the fast gas heating in pure nitrogen, as well as quenching of N( 2 D) atoms [3], ) ( N ) S ( N N ) D ( N 2 4 2 2 v , (1) which becomes extremely important at high gas temperatures. The temperature dependence of pooling reaction rate constant, which used in the model, was obtained based on the treatment of the experimental data [4].
Fig. 1. Temporal evolution of gas temperature in nitrogen plasma at 27 mbar. Points – are experimental data, curves – are the results of calculations with (curve 1) and without (curve 2) the heat release in reaction (1).
The work was partially supported by French National Research Agency, ANR (ASPEN Project), LabExPlas@Par and French-Russian international laboratory LIA KaPPA and RFBR project No 17-52- 16001.
[1] N. Lepikhin, A. Klochko, N. A. Popov and S. M. Starikovskaia, Plasma Sources Sci. Technol. 25 (2016) 045003. [2] N. A. Popov, J. Phys. D: 44 (2011) 285201. [3] B. Galvao et al. Phys. Chem. Lett. 4, (2013) 229 [4] G. Stancu, M. Janda, F. Kaddouri, D. Lacoste, C. O. Laux, J. Phys. Chem. A 114 (2010) 201. 10 295
XXXIII ICPIG, July 9-14, 2017, Estoril/Lisbon, Portugal
Control of charged species dynamics in atmospheric pressure plasmas using tailored voltage waveforms
A. R. Gibson 1, 2
, L. Alelyani 1 , S. Doyle 1 , J. Bredin 1 , J.-P Booth 2 , J. P. Dedrick 1 , T. Gans 1 and
D. O’Connell 1
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
Precise control of the chemical kinetics in atmospheric pressure plasma sources is crucial for their full potential to be realised in a range of applications. Radio-frequency plasmas driven by multiple driving frequencies offer an attractive route to achieve such control. In this work, we demonstrate wide-ranging control of charged species dynamics in He/N 2 plasma jets at atmospheric pressure using tailored voltage waveforms. Experimentally we employ using Phase Resolved Optical Emission Spectroscopy to measure the time and space resolved electron dynamics. Further insight into plasma control is obtained through comparison with one-dimensional fluid simulations.
Download 9.74 Mb. Do'stlaringiz bilan baham: 
|