1: \documentclass[twocolumn,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
2:
3: %\usepackage[latin1]{inputenc}
4: %\usepackage[active]{srcltx}
5:
6: \usepackage{graphicx}% Include figure files
7: \usepackage{dcolumn}% Align table columns on decimal point
8: \usepackage{bm}% bold math
9: \usepackage{epsfig}
10:
11: %\usepackage[active]{srcltx}
12:
13: \begin{document}
14:
15: \title[Electron Trapping]{Electron trapping by electric field reversal and Fermi mechanism}
16: \author{M\'{a}rio J. Pinheiro}
17: \address{Department of Physics and Centro de Fisica de Plasmas, Instituto Superior T\'{e}cnico,
18: Av. Rovisco Pais, 1049-001 Lisboa, Portugal}
19:
20: \email{mpinheiro@ist.utl.pt}
21:
22: \homepage{http://alfa.ist.utl.pt/~pinheiro}
23:
24:
25:
26: %\thanks{This work was done while I was in sabbatical leave at the University of Tennessee, in Knoxville. I would like to thank
27: %Prof. John Reece Roth for the critical reading of the manuscript.}
28: \pacs{52.10.+y, 52.20.Fs, 52.25.Dg, 52.25.Gj, 52.27.Aj}
29: %\subjclass{02.50.-r,03.40.Kf,03.65.-w} \keywords{Glow
30: %discharge,Fermi mechanism,Boltzmann equation,Field inversion}
31: \date{\today}
32:
33: \begin{abstract}
34: We investigate the existence of the electric field reversal in the
35: negative glow of a dc discharge, its location, the width of the
36: well trapping the electrons, the slow electrons scattering time,
37: and as well the trapping time. Based on a stress-energy tensor
38: analysis we show the inherent instability of the well. We suggest
39: that the Fermi mechanism is a possible process for pumping out
40: electrons from the through, and linking this phenomena with
41: electrostatic plasma instabilities. A power law distribution
42: function for trapped electrons is also obtained. Analytical
43: expressions are derived which can be used to calculate these
44: characteristics from geometrical dimensions and the operational
45: parameters of the discharge.
46:
47: \end{abstract}
48:
49: \pacs{52.10.+y, 52.20.Fs, 52.25.Dg, 52.25.Gj, 52.27.Aj}
50:
51: \bibliographystyle{apsrev}
52:
53: \maketitle
54:
55: %\pacs{52.10.+y,52.20.Fs,52.25.Dg,52.25.Gj,52.27.Aj}
56: %\date{\today}
57: %\dedicatory{}%
58: %\commby{}%
59:
60: % ----------------------------------------------------------------
61: \section{Introduction}
62:
63: The phenomena of field reversal of the axial electric field in the
64: negative glow of a dc discharge is of great importance, since the
65: fraction of ions returning to the cathode depends on its existence
66: and location. Technological application of gas discharges,
67: particularly to plasma display panels, needs a better knowledge of
68: the processes involved. The study of nonlocal phenomena in
69: electron kinetics of collisional gas discharge plasma have shown
70: that in the presence of field reversals the bulk electrons in the
71: cathode plasma are clearly separated in two groups of slow
72: electrons: trapped and free electrons\cite{Kolobov}. Trapped
73: electrons give no contribution to the current but represent the
74: majority of the electron population.
75:
76: The first field reversal it was shown qualitatively to be located
77: near the end of the negative glow (NG) where the plasma density
78: attains the greatest magnitude. If the discharge length is enough,
79: it appears a second field reversal on the boundary between the
80: Faraday dark space and the positive column. Also, it was shown in
81: the previously referred theoretical work that ions produced to the
82: left of this first reversal location move to the cathode by
83: ambipolar diffusion and ions generated to the right of this
84: location drift to the anode. For a review see also~\cite{Godyak}.
85: Those characteristic were experimentally observed by laser
86: optogalvanic spectroscopy~\cite{Gottscho}.
87:
88: Boeuf {\it et al}~\cite{Boeuf} with a simple fluid model gave an
89: analytical expression of the field reversal location which showed
90: to depend solely on the cathode sheath length, the gap length, and
91: the ionization relaxation length. They obtained as well a simple
92: analytical expression giving the fraction of ions returning to the
93: cathode and the magnitude of the plasma maximum density.
94:
95: In the present Letter we introduce a quite simple dielectric-like
96: model of a plasma-sheath system. This approach have been addressed
97: by other authors~\cite{Taillet69, Harmon76} to explain how the
98: electrical field inversion occurs at the interface between the
99: plasma sheath and the beginning of the negative glow. The aim of
100: this Letter is to obtain more information about the fundamental
101: properties related to field inversion phenomena in the frame of a
102: dielectric model. It is obtained a simple analytical dependence of
103: the axial location where field reversal occurs in terms of
104: macroscopic parameters. In addition, it is obtained the magnitude
105: of the minimum electric field inside the through, the trapped well
106: length, and the trapping time of the slow electrons into the well.
107: We emphasize in particular the description of the dielectric
108: behavior and do not contemplate plasma chemistry and
109: plasma-surface interactions.
110:
111: The analytical results hereby obtained could be useful for hybrid
112: fluid-particle models (e.g., Fiala {\it et al.}~\cite{Boeuf94}),
113: since simple criteria can be applied to accurately remove
114: electrons from the simulations.
115:
116: On the ground of the stress-energy tensor considerations it is
117: shown the inherent instability of the field inversion sheath. The
118: slow electrons distribution function is obtained assuming the
119: Fermi~\cite{Fermi} mechanism responsible for their acceleration
120: from the trapping well.
121:
122: \section{Theoretical model}
123:
124: Lets consider a plasma formed between two parallel-plate
125: electrodes due to an applied dc electric field. We assume a planar
126: geometry, but extension to cylindrical geometry is
127: straightforward. The applied voltage is $V_a$ and we assume the
128: cathode fall length is $l$ and the negative glow + eventually the
129: positive column extends over the length $l_0$, such that the total
130: length is $L=l + l_0$. We have
131: \begin{equation}\label{Eq1}
132: -V_a = l E_s + l_0 E_p,
133: \end{equation}
134: where $E_s$ and $E_p$ are, resp., the electric fields in the
135: sheath and NG (possibly including the positive column).
136:
137: At the end of the cathode sheath it must be verified the following
138: boundary condition by the displacement field $\mathbf{D}$
139: \begin{equation}\label{Eq2}
140: \mathbf{n}.(\mathbf{D}_p - \mathbf{D}_s) = \sigma.
141: \end{equation}
142: Here, $\sigma$ is the surface charge density accumulated at the
143: boundary surface and $\mathbf{n}$ is the normal to the surface. In
144: more explicit form,
145: \begin{equation}\label{Eq3}
146: \varepsilon_p E_p - \varepsilon_s E_s = \sigma.
147: \end{equation}
148: Here, $\varepsilon_s$ and $\varepsilon_p$ are, resp., the
149: electrical permittivity of the sheath and the positive column. We
150: have to solve the following algebraic system of equations
151: \begin{equation}\label{Eq5}
152: \begin{array}{cc}
153: l_0 E_p + l E_s & = - V_a, \\
154: \varepsilon_p E_p - \varepsilon_s E_s & = \sigma. \\
155: \end{array}
156: \end{equation}
157: They give the electric field strength in each region
158: \begin{equation}\label{Eq6}
159: \begin{array}{cc}
160: E_s = & -\frac{V_a}{L} \left(1-\alpha + \frac{l_o \sigma}{V_a \varepsilon_s}\right)\frac{1}{1-\frac{l\alpha}{L}}, \\
161: E_p = & -\frac{V_a}{L} \left(1-\frac{l \sigma}{V_a \varepsilon_s} \right) \frac{1}{1-\frac{l\alpha}{L}}.\\
162: \end{array}
163: \end{equation}
164: Here, we define
165: $\alpha=1-\frac{\varepsilon_p}{\varepsilon_s}=\frac{\omega_p^2}{\nu_{en}^2}$.
166: Recall that in DC case,
167: $\varepsilon_p=1-\frac{\omega_p^2}{\nu_{en}^2}$, and
168: $\varepsilon_s=\varepsilon_0$, with $\omega_p$ denoting the plasma
169: frequency and $\nu_{en}$ the electron-neutral collision frequency.
170: In fact, our assumption $\varepsilon_s=\varepsilon_0$ is plainly
171: justified, since experiments have shown the occurrence of a
172: significant gas heating and a corresponding gas density reduction
173: in the cathode fall region, mainly due to symmetric charge
174: exchanges processes which lead to an efficient conversion of
175: electrical energy to heavy-particle kinetic energy and thus to
176: heating~\cite{Hartog88}.
177:
178:
179: Two extreme cases can be considered: {\bf i}) $\omega_p
180: > \nu_{en}$, implying $\varepsilon_p < 0$, meaning that
181: $\tau_{coll} > \tau_{plasma}$, i.e, non-collisional regime
182: prevails; {\bf ii}) $\omega_p < \nu_{en}$, $\varepsilon_p > 0$,
183: and then $\tau_{coll}
184: > \tau_{plasma}$, i.e, collisional regime dominates.
185:
186: From the above Eqs.~\ref{Eq6} we estimate the field inversion
187: should occurs for the condition $1-\frac{l_o \alpha}{L}=0$, which
188: give the position on the axis where field inversion occurs:
189: \begin{equation}\label{Eq8}
190: \frac{l_o}{L} = \frac{\nu_{en}^2}{\omega_p^2}.
191: \end{equation}
192:
193: From Eq.~\ref{Eq8} we can resume a criteria for field reversal: it
194: only occurs in the non-collisional regime; by the contrary, in the
195: collisional regime and to the extent of validity of this simple
196: model, no field reversal will occur, since the slow electrons
197: scattering time inside the well is higher than the the well
198: lifetime, and collisions (in particular, coulombian collisions)
199: and trapping become competitive processes. A similar condition was
200: obtained in~\cite{Nishikawa69} when studying the effect of
201: electron trapping in ion-wave instability. Likewise, a
202: self-consistent analytic model~\cite{Kolobov} have shown that at
203: at sufficiently high pressure, field reversal is absent.
204:
205: Due to the accumulation of slow electrons after a distance
206: $\xi_c=L-l_0$, real charges accumulated on a surface separating
207: the cathode fall region from the negative glow. Naturally, it
208: appears polarization charges on each side of this surface and a
209: double layer is created with a surface charge $-\sigma_1' < 0$ on
210: the cathode side and $\sigma_2'$ on the anode side. But, $\sigma'
211: = (\mathbf{P} \cdot \mathbf{n})$,
212: $\mathbf{P}=\mathbf{\mathbf{\varepsilon}_0} \chi_e \mathbf{E}$
213: with $\varepsilon = \varepsilon_0 (1 + \chi_e)$, $\chi_e$ denoting
214: the dimensionless quantity called electric susceptibility. As the
215: electric displacement is the same everywhere, we have
216: $\mathbf{D}_0 = \mathbf{D}_1 = \mathbf{D}_2$. Thus, the residual
217: (true) surface charge in between is given by
218: \begin{equation}\label{Eq9}
219: \sigma = - \sigma_1' + \sigma_2'.
220: \end{equation}
221: After a straightforward but lengthy algebraic operation we obtain
222: \begin{equation}\label{Eq10}
223: \sigma = \varepsilon_p V_a \frac{B}{A},
224: \end{equation}
225: where
226: \begin{equation}\label{Eq11}
227: A = L \left( -1+ \frac{\varepsilon_0 -
228: \varepsilon_s}{\varepsilon_p} \right) + l\left( -
229: \frac{\varepsilon_p}{\varepsilon_s} +
230: \frac{\varepsilon_s}{\varepsilon_p} \right),
231: \end{equation}
232: and
233: \begin{equation}\label{Eq12}
234: B = \frac{\varepsilon_0 (\varepsilon_s - \varepsilon_p)}{
235: \varepsilon_s \varepsilon_p }.
236: \end{equation}
237: We can verify that $\sigma$ must be equal to
238: \begin{equation}\label{Eqsig}
239: \sigma = \alpha \frac{V_a \varepsilon_0}{2 l_0}.
240: \end{equation}
241: Considering that $\sigma = \varepsilon_0 \chi_e E$, we determine
242: the minimum value of the electric field at the reversal point:
243: \begin{equation}\label{Eq13}
244: E_m = \frac{\omega_p^2}{\nu_{en}^2} \frac{V_a}{2 l_0 \chi_e}.
245: \end{equation}
246:
247: Here, $\chi_e=\varepsilon_{rw}-1$, with $\varepsilon_{rw}$
248: designating the relative permittivity of the plasma trapped in the
249: well. From the above equation we can obtain a more practical
250: expression for the electrical field at its minimum strength
251: \begin{equation}\label{Eq14}
252: E_m = -
253: \frac{n_{ep}}{n_{ew}}\frac{\nu_{enw}^2}{\nu_{en}^2}\frac{V_a}{e
254: l_0} \approx - \frac{n_{ep}}{n_{ew}} \frac{T_{ew}}{T_{ep}}
255: \frac{V_a}{2 l_0}.
256: \end{equation}
257: The magnitude of the minimum electric field depends on the length
258: of the negative glow $l_0$. This also means that without NG there
259: is no place for field reversal, and also the bigger the length the
260: minor the electric field. The length of the negative glow can be
261: estimated by the free path length $l_0$ of the fastest electrons
262: possessing an energy equal to the cathode potential fall value
263: $eV_a$:
264: \begin{equation}\label{}
265: l_0 = \int_{0}^{eV_a} \frac{d w}{(N F(w))}.
266: \end{equation}
267: Here, $w$ is the electrons kinetic energy and $NF(w)$ is the
268: stopping power. For example, for He, it is estimated $p l_0=0.02
269: eV_a$ ~\cite{Kolobov} (in cm.Torr units, with $V_a$ in Volt). We
270: denote by $n_{ew}$ the density of trapped electrons and by
271: $T_{ew}$ their respective temperature. Altogether, $n_{ep}$ and
272: $T_{ep}$ are, resp., the electron density and electron temperature
273: in the negative glow region.
274:
275: By other side, we can estimate the true surface charge density
276: accumulated on the interface of the two regions by the expression
277: \begin{equation}\label{Eq15}
278: \sigma = \frac{Q}{A} = - \frac{n_{ep} e A \Delta \xi}{A}.
279: \end{equation}
280: Here, $Q$ is the total charge over the cross sectional area where
281: the current flows and $\Delta \xi$ is the width of the potential
282: well.
283:
284: \subsection{Instability and width of the potential well}
285:
286: From Eqs.~\ref{Eqsig} and ~\ref{Eq15} it is easily obtained the
287: trapping well width
288: \begin{equation}\label{Eq16}
289: \Delta \xi = - \frac{e V_a}{2 m l_0 \nu_{enw}^2}.
290: \end{equation}
291: It is expected that the potential trough should have a
292: characteristic width of the order in between the electron Debye
293: length ($\lambda_{De}=\sqrt{\frac{\varepsilon_0 kT_e}{n_e e^2}}$)
294: and the mean scattering length. Using Eq.~\ref{Eq16}, in a He
295: plasma and assuming $V_a=1$ kV, $l_0=1$ m and $\nu_{en}=1.85
296: \times 10^{9}$ s$^{-1}$ (with $T_e=0.03$ eV) at 1 Torr ($n=3.22
297: \times 10^{16}$ cm$^{-3}$) we estimate $\Delta \xi \approx 2.6
298: \times 10^{-3}$ cm, while the Debye length is $\lambda_{De}=2.4
299: \times 10^{-3}$ cm. So, our Eq.~\ref{Eq16} gives a good order of
300: magnitude for the potential width, which is expected to be in fact
301: of the same order of magnitude than the Debye length.
302:
303: Table I present the set of parameters used to obtain our
304: estimations. We give in Table II the estimate of the minimum
305: electric field attained inside the well. The first field reversal
306: at $\xi_c \approx l_{NG}$ corresponds to the maximum density
307: $n_{ew} \gg n_{ep}$~\cite{Boeuf,Tsendin2001}. So, the assumed
308: values for the ratio of electron temperatures and densities of the
309: trapped electrons and electrons on the NG are typical estimates.
310:
311:
312: \begin{table}
313: \centering
314: \caption{Data used for $E/p=100$ V$/$cm$/$Torr. Cross sections and electron
315: temperatures are
316: taken from Siglo Data base, CPAT and Kinema Software, http://www.Siglo-Kinema.com }\label{Table1}
317: \begin{tabular}{|c|c|c|}
318: \hline
319: % after \\: \hline or \cline{col1-col2} \cline{col3-col4} ...
320: Gas & $T_e$ (eV) & $\sigma$ ($10^{-16}$ cm$^{2}$)\\
321: \hline
322: Ar & 8 & 4.0 \\
323: He & 35 & 2.0 \\
324: O$_2$ & 6 & 4.5 \\
325: N$_2$ & 4 & 9.0 \\
326: H$_2$ & 8 & 6.0 \\
327: \hline
328: \end{tabular}
329: \end{table}
330:
331: It can be shown that there is no finite configuration of fields
332: and plasma that can be in equilibrium without some external
333: stress~\cite{Longmire}. Consequently, this trough is dim to be
334: unstable and burst electrons periodically (or in a chaotic
335: process), releasing the trapped electrons to the main plasma. This
336: phenomena produces local perturbation in the ionization rate and
337: the electric field giving rise to ionization waves (striations).
338: In the next section, we will calculate the time of trapping with a
339: simple Brownian model.
340:
341:
342: \begin{table}\label{Table2}
343: \centering
344: \caption{Minimum electric field at reversal point and well width.
345: Conditions: He gas, $p=1$ Torr, $l_0=20$ cm, $V_a=1$ kV, $\frac{T_{ew}}{T_{ep}}=0.1$, $\frac{n_{ew}}{n_{ep}}=10$.}\label{Table1}
346: \begin{tabular}{|c|c|}
347: \hline % after \\: \hline or \cline{col1-col2} \cline{col3-col4} ...
348: $E_m$ (V.cm$^{-1}$) & $\Delta \xi$ (cm) \\
349: \hline
350: $\lesssim -2.5$ & $2.6 \times 10^{-3}$ \\
351: \hline
352: \end{tabular}
353: \end{table}
354:
355: From Eq.~\ref{Eq8} we calculate the cathode fall length for some
356: gases. For this purpose we took He and H$_2$ data as reference for
357: atomic and molecular gases, resp. The orders of magnitude are the
358: same, with the exception of Ar. Due to Ramsauer effect direct
359: comparison is difficult.
360:
361: In Table III it is shown a comparison of the experimental cathode
362: fall distances to the theoretical prediction, as given by
363: Eq.~\ref{Eq16}. Taking into account the limitations of this model
364: these estimates are well consistent with experimental
365: data~\cite{Brown59}.
366:
367: \begin{table}
368: \centering
369: \caption{Comparison between theoretical and experimental cathode fall distance at p=1 Torr,
370: $E/p$=100 V$/$cm$/$Torr. Experimental data are collected from Ref.~\cite{Brown59}. }\label{Table2}
371: \begin{tabular}{|c|c|c|}
372: \hline
373: % after \\: \hline or \cline{col1-col2} \cline{col3-col4} ...
374: Gas & $\xi_c^{teo}$ (cm) & $\xi_c^{exp}$ (cm) \\
375: \hline
376: Ar & 7.40 & 0.29 (Al) \\
377: He & 1.32 & 1.32 (Al) \\
378: $H_2$ & 0.80 & 0.80 (Cu) \\
379: $N_2$ & 0.45 & 0.31 (Al) \\
380: $Ne$ & 0.80 & 0.64 (Al) \\
381: $O_2$ & 0.30 & 0.24 (Al) \\
382: \hline
383: \end{tabular}
384: \end{table}
385:
386:
387: \subsection{Lifetime of a slow electron in the potential well}
388:
389: The trapped electrons most probably diffuse inside the well with a
390: characteristic time much shorter than the lifetime of the through.
391: Trapping can be avoided by Coulomb collisions~\cite{Nishikawa69}
392: or by the ion-wave instability, both probably one outcome of the
393: stress energy unbalance as previously mentioned. We consider a
394: simple Brownian motion model for the slow electrons to obtain the
395: scattering time $\tau$, and the lifetime T of the well. A
396: Fermi-like model will allow us to obtain the slow electron energy
397: distribution function.
398:
399: Considering the slow electron jiggling within the well, the
400: estimated scattering time is
401: \begin{equation}\label{scat1}
402: \tau = \frac{(\Delta \xi)^2}{ \mathcal{D}_e }.
403: \end{equation}
404: Here, $\mathcal{D}_e$ is the electron diffusion coefficient at
405: thermal velocities.
406:
407:
408:
409: \begin{table*}
410: \caption{Scattering time and trapping time in the well. The parameters are:
411: $E/N=100$ Td, $T_g=300$ K, $V_a=1$ kV and $l_0=0.1$ m.}\label{Table 4}
412: \begin{ruledtabular}
413: \begin{tabular}{cccccc}
414: % \hline
415: % after \\: \hline or \cline{col1-col2} \cline{col3-col4} ...
416: Gas & $\mathcal{D}_e$ (cm$^2$.s$^{-1}$)\footnote{Data obtained through resolution of the homogeneous electron Boltzmann equation
417: with two term expansion of the distribution function in spherical harmonics, M. J. Pinheiro and J. Loureiro,
418: J. Phys. D.: Appl. Phys. {\bf 35} 1 (2002) } & $\nu_{enw} (s^{-1})$ \footnote{Same remark as in ${}^a$} & $\Delta \xi (cm)$ & $\tau$ (s) & T (s) \\
419: \hline
420: Ar & $2.52 \times 10^6$ & $8.10 \times 10^9$ & $1.34 \times 10^{-3}$ & $7.10 \times 10^{-13}$ & $3.97 \times 10^{-5}$ \\
421: He & $5.99 \times 10^6$ & $2.39 \times 10^9$ & $1.54 \times 10^{-2}$ & $3.95 \times 10^{-11}$ & $1.70 \times 10^{-5}$\\
422: N$_2$ & $6.11 \times 10^5$ & $6.15 \times 10^9$ & $2.32 \times 10^{-3}$ & $8.81 \times 10^{-12}$ &$1.64 \times 10^{-4}$ \\
423: CO$_2$ & $1.70 \times 10^6$ & $3.60 \times 10^9$ & $6.78 \times
424: 10^{-3}$& $2.70 \times 10^{-11}$ & $5.90 \times 10^{-5}$ \\
425: % \hline
426: \end{tabular}
427: \end{ruledtabular}
428: \end{table*}
429:
430:
431:
432: The fluctuations arising in the plasma are due to the breaking of
433: the well and we can estimate the amplitude of the fluctuating
434: field by means of Eq.~\ref{Eq14}. We obtain
435: \begin{equation}\label{}
436: \delta E_m =
437: \frac{n_{ep}}{n_{ew}}\frac{\nu_{enw}^2}{\nu_{en}^2}\frac{V_a}{e
438: l_0^2} \Delta \xi.
439: \end{equation}
440:
441: Then, we have
442: \begin{equation}\label{}
443: \mathcal{E}_c = \frac{\delta E_m}{E_m} = \frac{\Delta \xi}{l_0}.
444: \end{equation}
445:
446: In Table IV we summarize scattering and trapping times for a few
447: gases.
448:
449: \subsection{Power-law slow electrons distribution function}
450:
451: As slow electrons are trapped by the electric field inversion,
452: some process must be at work to pull them out from the well. We
453: suggest that fluctuations of the electric field in the plasma
454: (with order of magnitude of $\mathcal{E}_c$)act over electrons
455: giving energy to the slow ones, which collide with those
456: irregularities as with heavy particles. From this mechanism it
457: results a gain of energy as well a loss. This model was first
458: advanced by E. Fermi~\cite{Fermi} when developing a theory of the
459: origin of cosmic radiation. We shall focus here on the rate at
460: which energy is acquired.
461:
462: The average energy gain per collision by the trapped electrons (in
463: order of magnitude) is given by
464: \begin{equation}\label{}
465: \Delta w = \overline{U} w(t),
466: \end{equation}
467: with $\overline{U}\cong \mathcal{E}_c^2$ and where $w$ is their
468: kinetic energy. After $N$ collisions the electrons energy will be
469: \begin{equation}\label{}
470: w(t) = \varepsilon_{t} \exp \left( \frac{\overline{U}t}{\tau}
471: \right),
472: \end{equation}
473: with $\varepsilon_t$ being their thermal energy, typical of slow
474: electrons.
475: \begin{figure}
476: % \usepackage{graphicx}
477: \includegraphics[width=3 in, height=4.5 in]{Function1.EPS}\\
478: \caption{Slow electrons distribution function vs. energy, for the same conditions as presented in Table IV. Solid curve: Ar, broken curve: N$_2$.}\label{}
479: \end{figure}
480: The time between scattering collisions is $\tau$. Assuming a
481: Poisson distribution $P(t)$ for electrons escaping from the
482: trapping, then we state
483: \begin{equation}\label{5}
484: P(t)=\exp(-t/\tau)dt/T.
485: \end{equation}
486: The probability distribution of the energy gained is a function of
487: one random variable (the energy), such as
488: \begin{equation}\label{6}
489: f_w(w)d w = P\{ w<\bar{w}<w+dw \}.
490: \end{equation}
491: This density $f_w(w)$ can be determined in terms of the density
492: P(t). Denoting by $t_1=T$ the real root of the equation
493: $w=w(t_1=T)$, then it can be readily shown that slow electrons
494: obey in fact to the following power-law distribution function
495: \begin{equation}\label{7}
496: f_w(w) d w = \frac{\tau}{\bar{U}T}
497: \varepsilon_{t}^{\frac{\tau}{\bar{U}T}} \frac{d
498: w}{w^{1+\tau/\bar{U}T}}.
499: \end{equation}
500: Like many man made and naturally occurring phenomena (e.g.,
501: earthquakes magnitude, distribution of income), it is expected the
502: trapped electron distribution function to be a power-law (see
503: Eq.~\ref{7}), hence $1 + \frac{\tau}{\mathcal{E}_c^2 T} = n$, with
504: $n=2 \div 4$ as a reasonable guess. Hence, we estimate the
505: trapping time to be
506: \begin{equation}\label{traptime}
507: T \approx \frac{\tau}{\mathcal{E}_c^2 n}.
508: \end{equation}
509:
510: Fig.1 shows the slow electrons distribution function pumped out
511: from the well for two cases: Ar (solid curve), and N$_2$ (broken
512: curve). It was chosen a power exponent $n=2$. Those distributions
513: show that the higher confining time is associated with less slow
514: electrons present in the well. When the width of the well
515: increases (from solid to broken curve) the scattering time become
516: longer, and as well the confining time, due to a decrease of the
517: relative number of slow electrons per given energy. This mechanism
518: of pumping out of slow (trapped) electrons from the well can
519: possibly explains the generation of electrostatic plasma
520: instabilities.
521:
522: Note that the trapping time is, in fact, proportional to the
523: length of the NG and inversely proportional to the electrons
524: diffusion coefficient at thermal energies:
525: \begin{equation}\label{}
526: T \approx \frac{l_0^2}{\mathcal{D}_e}.
527: \end{equation}
528: The survival frequency of trapped electrons is $\nu_t=1/T$. As the
529: electrons diffusion coefficient are typically higher in atomic
530: gases, it is natural to expect plasma instabilities and waves with
531: higher frequencies in atomic gases. This result is in agreement
532: with a kinetic analysis of instabilities in microwave
533: discharges~\cite{Tatarova1}. In addition, the length of the NG
534: will influence the magnitude of the frequencies registered by the
535: instabilities, since wavelengths have more or less space to
536: build-up. Table~\ref{Table 4} summarizes the previous results for
537: some atomic and molecular gases. The transport parameters used
538: therefor where calculated by solving the electron Boltzmann
539: equation, under the two-term approximation, in a steady-state
540: Townsend discharge~\cite{Pinheiro}
541:
542: \section{Conclusion}
543:
544: We have shown in the framework of a simple dielectric model that
545: the magnitude of the minimum electric field (on the edge of the
546: negative glow) depends directly on the applied voltage and is
547: inversely proportional to the NG length.
548:
549: The width of the well trapping the slow electrons is directly
550: dependent on the applied electric field and is inversely
551: proportional to the square of the electron-neutral collision
552: frequency for slow electrons. It is, as well, inversely
553: proportional to the NG length, and has typically the extension of
554: a Debye length. We state that for typical conditions of a
555: low-pressure glow-discharge, field reversal occurs whenever
556: $\omega_p
557: > \nu_{en}$, due to a lack of collisions necessary to pump out electrons from the well.
558: Furthermore, the analytical expressions obtained for the
559: scattering and trapping time of the slow electrons are potentially
560: useful in hybrid fluid-particle plasma modelling.
561:
562:
563: \begin{thebibliography}{1}
564:
565: \bibitem{Kolobov} V. I. Kolobov and L. D. Tsendin, Phys. Rev.
566: A {\bf 46}(12), 7837 (1992)
567:
568: \bibitem{Godyak} Vladimir I. Kolobov and Valery A. Godyak,
569: IEEE Transactions on Plasma Science, {\bf 23}(4), 503 (1995)
570:
571: \bibitem{Gottscho} Richard A. Gottscho, Annette Mitchell,
572: Geoffrey R. Scheller, Yin-Yee Chan, David B. Graves, Phys.
573: Rev. A, {\bf 40}(1), 6407 (1989)
574:
575: \bibitem{Boeuf} J. P. Boeuf and L. C. Pitchford, J. Phys. D:
576: Appl. Phys. {\bf 28}, 2083 (1995)
577:
578: \bibitem{Taillet69} J. Taillet, Am. J. Phys. {\bf 37}, 423
579: (1969)
580:
581: \bibitem{Harmon76} Gerald S. Harmon, Am. J. Phys. {\bf 44}(9),
582: 869 (1976)
583:
584: \bibitem{Boeuf94} A. Fiala, L. C. Pitchford, and J. P. Boeuf,
585: Phys. Rev. E {\bf 49} (6), 5607 (1994)
586:
587: \bibitem{Fermi} E. Fermi, Phys. Rev. {\bf 75} (8), 1169 (1949)
588:
589: \bibitem{Hartog88} E. A. Den Hartog, D. A. Doughty, and J. E.
590: Lawler, Phys. Rev. A, {\bf 38} (5), 2471 (1988)
591:
592: \bibitem{Nishikawa69} K. Nishikawa and Ching-Sheng Wu,
593: Phys. Rev. Lett. {\bf 23}(18), 1020 (1969)
594:
595: \bibitem{Tsendin2001} A. A. Kudryatsev and L.D. Tsendin,
596: Technical Physics Letters, {\bf 27}(4), 284 (2001)
597:
598: \bibitem{Brown59} Sanborn C. Brown, {\it Basic data of plasma
599: physics} (The MIT Press, Cambridge, 1959)
600:
601: \bibitem{Longmire} Conrad L. Longmire, {\it Elementary Plasma,
602: Physics} (John Wiley $\&$ Sons, New
603: York, 1963), Section 3.7
604:
605: \bibitem{Tatarova1} A. Shivarova, E. Tatarova, and V.
606: Angelova, J. Phys. D: Appl. Phys. {\bf 21} 1605 (1988)
607:
608: \bibitem{Pinheiro} M. J. Pinheiro and J. Loureiro, J. Phys. D: Appl. Phys. {\bf 35}, 3077 (2002)
609:
610:
611: \end{thebibliography}
612:
613:
614: % ----------------------------------------------------------------
615: %\INPUT{Xbib.bib} % For Gather Purpose Only
616: %\INPUT{PhysRev1.bbl} % For Gather Purpose Only
617: %\bibliographystyle{amsplain}
618: %\bibliography{xbib}
619:
620: \end{document}
621: % ----------------------------------------------------------------
622: