1: %********************************************************************
2: \documentclass[epj]{svjour}
3: \usepackage{epsfig,epsf}
4: %---------------------------------------------------------------------
5: \begin{document}
6:
7: \title{Resonant fluxon transmission through impurities}
8: \author{ Yaroslav Zolotaryuk}
9: %\address{
10: \institute{
11: Bogolyubov Institute for Theoretical Physics,
12: National Academy of Sciences of Ukraine,
13: vul. Metrologichna 14B,
14: 03680 Kyiv, Ukraine,
15: \email{yzolo@bitp.kiev.ua}
16: }
17: \date{Received: date / Revised version: date}
18: \abstract{
19: Fluxon transmission through several impurities of different
20: strength and type (i.e., microshorts and microresistors),
21: placed in a long Josephson junction is investigated.
22: Threshold pinning current on the impurities is computed
23: as a function of the distance between them, their amplitudes and
24: the dissipation parameter. It is shown that in the case
25: of consequently placed microshorts or microresistors, the threshold
26: pinning current exhibits a clear minimum as a function of the
27: distance between the impurities. In the case of a microresistor,
28: followed by a microshort, an opposite phenomenon is observed, namely
29: the threshold pinning current exhibits maximum as a function of
30: the distance between the impurities.
31: \PACS{
32: {03.75.Lm}{Josephson vortices} \and
33: {05.45.Yv}{Solitons} \and
34: {74.50.+r}{Josephson effect}
35: }}
36: \maketitle
37:
38:
39: %######################################################################
40: %######################################################################
41: \section{Introduction and the background}
42: %######################################################################
43: %######################################################################
44:
45: The dynamics of magnetic flux propagation in a long Josephson
46: junction (LJJ) is a subject of increasing theoretical and
47: practical interest \cite{barone82,u98pd}. Magnetic flux quantum
48: in a LJJ is a soliton (also known as {\it fluxon})
49: governed by the well-known sine-Gordon (SG) equation.
50: A convenient way to prepare a junction with the required
51: properties is to install various inhomogeneities into it.
52: Up to now substantial work has been devoted to the study of the
53: fluxon motion in the LJJs with point-like impurities. The interaction
54: of a fluxon with a single impurity became a textbook
55: example \cite{ms78pra}.
56:
57: On the other hand, the phenomenon of resonant tunneling of an
58: electron through a
59: double-well structure is well-known in quantum mechanics \cite{restun}.
60: A natural question arises: what is an analog of the quantum-mechanical
61: resonant tunneling in the fluxon dynamics? Resonant soliton
62: transmission has been investigated in detail for non-dissipative
63: systems \cite{ksckv92jpa,zkv94jpsj} and complex resonant
64: behaviour has been reported. However, fluxon dynamics in a LJJ cannot
65: be considered without taking into account dissipative effects,
66: which are a consequence of the normal electron tunneling across
67: the insulating barrier. As a result, transmission in a LJJ with
68: constant bias and dissipation can yield only two scenarios: fluxon
69: transmission or fluxon pinning on the impurities. And,
70: consequently, the transmission ratio
71: can attain only two values: zero or unity. Therefore the attention has to
72: be turned toward other characteristic quantities, especially the minimal
73: bias, necessary for the fluxon pinning on impurities.
74:
75: The present paper aims to investigate fluxon transmission through
76: several (two or more) point-like impurities: microshorts, microresistors
77: or a combination of both. Of particular interest is dependence of the
78: threshold pinning current on the distance between the impurities and their
79: amplitudes.
80:
81: The paper is organized as follows. In Section \ref{s2} we present
82: the model and the basic equations of motion. In the
83: next section we describe the methods of the analysis of the
84: equations and motion and study the
85: fluxon transmission through two microshorts, two microresistors
86: and a microshort and a microresistor as a function of their
87: amplitudes and distance between them. Discussion of the obtained
88: results and final remarks are given in Sec. \ref{s4}.
89:
90:
91: %######################################################################
92: %######################################################################
93: \section{The model}
94: \label{s2}
95: %######################################################################
96: %######################################################################
97:
98: We consider the long Josephson junction (LJJ) subjected to the external
99: time-independent bias. The main dynamical variable is the difference
100: between the phases $\theta_2(x,t)-\theta_1(x,t)=\phi(x,t)$ of the
101: macroscopic wave functions of the the superconducting layers of the
102: junction. The time evolution of the phase difference is governed by
103: the perturbed sine-Gordon (SG) equation:
104: %---------------------------------------------------------------------
105: \begin{eqnarray}
106: \nonumber
107: &&\phi_{tt}-\phi_{xx}+ \sin \phi = \epsilon f [\phi, \phi_t; x]~, \\
108: && \epsilon f [\phi, \phi_t; x]\doteq -\alpha \phi_t -
109: \gamma-\sin \phi \sum_{n=1}^{m} \mu_n \delta (x-a_m).
110: \label{1}
111: \end{eqnarray}
112: %---------------------------------------------------------------------
113: In this dimensionless equation spacial variable $x$ is normalized to
114: the Josephson penetration depth $\lambda_J$, the temporal variable $t$
115: is normalized to the inverse
116: Josephson plasma frequency $\omega_J^{-1}$\cite{parameters}.
117: Here the bias current $\gamma$ is normalized to the critical Josephson
118: current of the junction and the dimensionless parameter $\alpha$
119: describes dissipation. It is supposed that there are $N$ impurities
120: in this junction,
121: positioned at the points $x=a_n$, $n=1,2,\ldots,N$,
122: $a_1\equiv 0<a_2<...<a_m$, with $\mu_n$ being ``strength'' or amplitude
123: of the $n$th impurity. The impurity is a microshort if $\mu_n>0$ and a
124: microresistor if $\mu_n<0$.
125:
126:
127: %######################################################################
128: %######################################################################
129: \section{Fluxon transmission}
130: \label{s3}
131: %######################################################################
132: %######################################################################
133:
134: A standard tool for analyzing the fluxon dynamics in Josephson
135: junctions is the McLaughlin-Scott perturbation theory \cite{ms78pra}.
136: Also, direct numerical
137: integration\footnote{
138: In the numerical simulations, the space will be discretized
139: as $x \to nh$, so that the continuous variable
140: $\phi(x,t)\simeq \phi(nh,t)$ becomes a discrete set of
141: variables $\phi_n(t)$, and the second space
142: derivative becomes
143: $\phi_{xx}(x,t) \simeq [\phi_{n+1}(t)-2\phi_n(t)+\phi_{n-1}(t)]/h^2$.
144: The resulting set of the second order ODEs on $\phi_n(t)$
145: will be solved using the 4th order Runge-Kutta scheme.
146: The delta function is approximated as
147: $\delta(x)\simeq \delta_{n,0}/h$ where $\delta_{m,n}$ is
148: Kronecker's $\delta$ symbol.
149: }
150: of the perturbed SG equation
151: (\ref{1}) will be performed to check the validity of the analytical
152: approximation. We are going to solve the problem for the idealized case
153: of an infinite junction with free ends boundary conditions,
154: however, in actual simulation a sample with length that significantly
155: exceeds the fluxon size will be used.
156:
157: \subsection{Perturbation theory and collective coordinates}
158: \label{s3.0}
159:
160: Using the perturbation theory, one obtains in the first order the
161: evolution equations for the fluxon
162: parameters, i.e., its center of mass $X$ and fluxon velocity $v$:
163: %----------------------------------------------------------------------
164: \begin{eqnarray}
165: \nonumber
166: &&\dot v = \frac{\pi \gamma}{4}(1-v^2)^{3/2}-\alpha v(1-v^2)+\\
167: && +\frac{1-v^2}{2}
168: \label{2}
169: \sum_{n=1}^N \mu_n g [X-a(n-1),v]~,\\
170: &&\dot X = v - \frac{v}{2}
171: \label{3}
172: \sum_{n=1}^N \mu_n [X-a(n-1)]g[X-a(n-1),v],\\
173: &&g(X,v)\doteq \frac{\tanh{(X/\sqrt{1-v^2})}}{\cosh^2{(X/\sqrt{1-v^2})}}~.
174: \nonumber
175: \end{eqnarray}
176: %----------------------------------------------------------------------
177: For the sake of simplicity in the following only equidistant impurities
178: will be considered, i.e., $a_n \equiv a$, $n=1,2,\ldots,N$.
179: Also, only positive values of bias $\gamma$ will be considered.
180: The case of one impurity ($N=1$, $\mu_1\equiv \mu$) has been discussed
181: in detail in \cite{ms78pra,kmn88jetp}. There exist two characteristic values of
182: the bias current, $\gamma_c \equiv 4\sqrt{3}\mu/(9\pi)$ and
183: $\gamma_{thr}$, $\gamma_c > \gamma_{thr}$. If $\gamma > \gamma_c$, the
184: pinning on the impurity is not possible and only one attractor that
185: corresponds to fluxon propagation does exist. In the interval
186: $\gamma_{thr}< \gamma < \gamma_c$ two attractors exist: one corresponds
187: to fluxon pinning on the microshort and another one to fluxon
188: propagation. If $\gamma < \gamma_{thr}$, the only possible regime is
189: fluxon pinning on the impurity. It has been
190: shown \cite{ms78pra,kmn88jetp} that there exists a threshold value
191: of the dc bias, which can be approximated as
192: %----------------------------------------------------------------------
193: \begin{equation}
194: \gamma_{thr}=\frac{\alpha}{\pi}\sqrt{8 \mu + \mu^2}
195: ~\left [1-2\alpha \ln 2 \right ]~.
196: \label{4a}
197: \end{equation}
198: %----------------------------------------------------------------------
199: In the case of one microresistor $\mu<0$, the threshold bias has been
200: defined in \cite{kmn88jetp} as
201: %----------------------------------------------------------------------
202: \begin{equation}
203: \gamma_{thr}=2\frac{\alpha}{\pi} \left [2\sqrt{\sqrt{2|\mu|}\pi\alpha}
204: +\frac{9}{2}\alpha \ln (\alpha \sqrt{2/|\mu|})\right ]~.
205: \label{4b}
206: \end{equation}
207: %----------------------------------------------------------------------
208: In the non-relativistic limit ($v \ll 1 $) the system (\ref{2})-(\ref{3})
209: can be rewritten
210: as a Newtonian second order ODE for the particle of mass $8$
211: (see Refs. \cite{ms78pra,kmn88jetp}):
212: %----------------------------------------------------------------------
213: \begin{eqnarray}
214: \label{6a}
215: &&8 \ddot X+ 8\alpha \dot X = - \frac{\partial U}{\partial X}~,\\
216: \label{6b}
217: && U(X)=-2 \pi \gamma X+U_0(X) \equiv -2 \pi \gamma X+\\
218: \nonumber
219: &&+2\sum_{n=1}^m \frac{\mu_n}{\cosh^2 {[X-a(n-1)]}}~,~~ \gamma \geq 0~.
220: \end{eqnarray}
221: %----------------------------------------------------------------------
222: In the case of strongly separated impurities
223: ($a \gg 1$), the potential $U(X)$ has $2N$ extrema [they approximately
224: coincide with the fixed points $X=X_{2k}$, $k=1,2,\ldots,N$, of equations
225: (\ref{2})-(\ref{3})]
226: where each pair (a minimum and a maximum) is associated with a certain
227: impurity. If there is an impurity at $X=a(k-1)$,
228: the minimum at $X=X_{2k-1}$ always comes
229: before the maximum at $X=X_{2k}$. Microshorts are repelling
230: impurities, thus the fluxon that arrives from $X=-\infty$
231: decelerates when approaching it and accelerates
232: after passing the impurity until the fluxon velocity reaches
233: the equilibrium value
234: %----------------------------------------------------------------------
235: \begin{equation}
236: v_\infty=\left [ 1+ \left( \frac{4\alpha}{\pi\gamma}\right )\right ]^{-1/2}~.
237: \label{7}
238: \end{equation}
239: %----------------------------------------------------------------------
240: Microresitors are
241: attractive impurities, and, as a result, the fluxon accelerates before
242: approaching the microresistor
243: and slows down to the equilibrium velocity (\ref{7}) after being
244: released from it. Decrease of the distances between the impurities
245: $a$ causes disappearance of some of the extrema via inverse pitchfork
246: bifurcations. The systematic phase plane analysis of equations
247: (\ref{2})-(\ref{3}) for the case of two
248: microshorts has been performed in \cite{bp90pla} for the SG
249: equation and in \cite{pb94pla} for the double SG
250: equation. In those papers the behaviour of the fixed points
251: of the system (\ref{2})-(\ref{3}) has been studied as a function
252: of the distance between them, $a$.
253:
254: Our aim is to determine the threshold current
255: $\gamma_{thr}=\gamma_{thr}(a,\{\mu_n \}_1^N;N)$ as a function of
256: the distance between impurities and their amplitudes. In the
257: case of one impurity, $\gamma_{thr}(\mu;1)$ obviously does not
258: depend on the distance $a$. It is described approximately by
259: equations (\ref{4a})
260: and (\ref{4b}) for the microshort and microresistor, respectively.
261: Some general statement can be made before one proceeds
262: to specific cases. Two important limits should be mentioned.
263: One case corresponds to impurities being separated
264: by the distance much larger the fluxon size. Then the transmission will be
265: governed by the fluxon interaction with each individual impurity. In
266: the opposite limit ($a \rightarrow 0$) the power of all impurities
267: adds up. The effect of the both limits on the threshold current can be
268: written as follows
269: %----------------------------------------------------------------------
270: \begin{equation}
271: \label{6}
272: \gamma_{thr}(a,\{\mu_n \}_1^N;N) = \left \{
273: \begin{array}{c}
274: \gamma_{thr}\left (\sum_{n=1}^N \mu_n;1 \right ),~a \to 0, \\
275: \max_{1\le n\le N}[ \gamma_{thr}(\mu_n;1) ],~a \to \infty.
276: \end{array}
277: \right .
278: \end{equation}
279: %----------------------------------------------------------------------
280:
281: In the subsections below the transmission through impurities of different
282: polarities (e.g, $\mu_n<0$ and $\mu_n>0$) will be considered.
283: It appears that for $N \ge 2$ the analytical treatment of equations
284: (\ref{2})-(\ref{3}) is virtually not possible even in the
285: non-relativistic case, especially when none of the limits,
286: described by the equation (\ref{6}), hold. Therefore equations
287: (\ref{2})-(\ref{3}) are going to be solved numerically.
288:
289:
290:
291: %######################################################################
292: \subsection{Transmission through two microshorts}
293: \label{s3.1}
294: %######################################################################
295:
296: Consider first the case of two microshorts ($N=2$, $\mu_{1,2}>0$).
297: The problem is tackled in the following way. The fluxon approaches
298: the system of two microshorts from $X=-\infty$ with the equilibrium
299: velocity $v_\infty$, given by equation (\ref{7}). Evolution of the
300: system (\ref{2})-(\ref{3}) on the phase plane $(X,v)$ is shown
301: in Figure \ref{fig1}. Depending on the strength
302: of the bias, three scenarios are possible: trapping on the first
303: microshort (curve 1);
304: trapping on the second microshort, if the external bias is
305: a bit larger (curve 2); or transmission (curve 3).
306: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
307: %
308: %-------------------Figure 1-------------------------------------------
309: \begin{figure}[htb]
310: \centerline{\epsfig{file=fig1.eps,width=2.8in,angle=-90}}
311: \caption{Phase space trajectories of equation (\ref{2}) for $N=2$
312: microshorts with
313: $\mu_1=0.4$, $\mu_2=0.6$ and distance $a=3$ between them. Curve $1$
314: corresponds to $\gamma=0.04$, curve $2$ - to $\gamma=0.05$ and
315: curve $3$ - to $\gamma=0.06$. Blue curve shows results of
316: direct integration of the equation (\ref{1}) at $\gamma=0.06$
317: the using Runge-Kutta method. The fluxon center of mass
318: is defined as $X(t)=\int_{-\infty}^{+\infty} x \phi_x dx $ and
319: its velocity as $v(t)=dX(t)/dt$.
320: Dissipation for all cases is $\alpha=0.1$.
321: The inset shows the effective potential $U(X)$ [equation (\ref{6b})]
322: for $\mu_1=0.4,\mu_2=0.6$, $\gamma=0.05$, $a=1$ (curve 1),
323: $a=1.7$ (curve 2), $a=2.5$ (curve 3) and $a=4$ (curve 4).
324: Red vertical bars denote locations of the microshorts and the lengths
325: of the bars is proportional to the microshort amplitudes, $\mu_{1,2}$.
326: }
327: \label{fig1}
328: \end{figure}
329: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
330: If the microshorts are too close to each other, trapping on the second
331: microshort does not happen (see references \cite{bp90pla,pb94pla} for
332: details). We note that direct numerical simulations of the perturbed SG
333: equation (\ref{1}) (curve 4 of Figure \ref{1}) are in
334: good correspondence with the trajectories of the system (\ref{2}). The
335: oscillations after the collision with the microshort can be attributed
336: to the fluxon radiation (not accounted by the first order perturbation
337: theory) and errors in determination of the fluxon center.
338:
339: The systematic evaluation of the threshold current $\gamma_{thr}$ as
340: a function of the distance $a$ for different values of $\mu_1$ and
341: $\mu_2$ is shown in Figure \ref{fig2}. The resonant nature of the
342: dependence of $\gamma_{thr}$ on $a$ for $\mu_1<\mu_2$ can be observed
343: clearly. While in the respective limiting cases it satisfies
344: equation (\ref{6}), a resonant value $a=a_r$ appears, at which the
345: threshold current attains its minimal value.
346: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
347: %
348: %-------------------Figure 2-------------------------------------------
349: \begin{figure}[htb]
350: \centerline{\epsfig{file=fig2.eps,width=2.8in,angle=-90}}
351: %\end{center}
352: \caption{Threshold pinning current as a function of the distance
353: between the microshorts for $\mu_1=0.6$, $\mu_2=0.4$ ($\Delta$);
354: $\mu_1=\mu_2=0.5$ ($\circ$);
355: $\mu_1=0.4$, $\mu_2=0.6$ ($+$); $\mu_1=0.3$, $\mu_2=0.7$ ($\times$);
356: $\mu_1=0.2$, $\mu_2=0.8$ ($\nabla$).
357: Results, obtained via direct numerical integration of the equations
358: of motion for $\mu_1=0.4$, $\mu_2=0.6$ are shown by $\oplus$.}
359: \label{fig2}
360: \end{figure}
361: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
362:
363: The explanation of the resonant transmission can be done on the following
364: qualitative argument. The analysis of the phase portraits in
365: Figure \ref{1} shows that after being released from the microshort,
366: the fluxon
367: accelerates in order to regain its equilibrium velocity $v_\infty$.
368: This
369: acceleration occurs in such a way that for some short interval the
370: fluxon
371: velocity exceeds the equilibrium value $v_\infty$. Therefore the fluxon
372: has kinetic energy, which is larger than it
373: was while approaching the microshort from $X=-\infty$, and
374: consequently it has enough energy to pass the microshort with the
375: amplitude larger than $\mu_1$. Obviously, the best transmission would
376: take place if $a$ slightly exceeds $|X_{2}-X_1|$. The
377: estimation of the
378: resonant distance $a_r$ can be made from the analysis of the fluxon
379: dynamics in the non-relativistic limit, given by
380: equations (\ref{6a})-(\ref{6b}).
381: According to these equations the fluxon can be compared to the
382: particle that slides down along the potential
383: $U(X)=U(X \to \pm \infty) \sim - 2\pi \gamma X$. Depending on the value
384: of $\gamma$, it can be trapped in one of the wells of this potential
385: (shown in the inset of Figure \ref{fig1}). If the distance between
386: microshorts is small enough, it can be considered as one microshort with
387: the renormalized
388: strength ${\bar \mu}(a)=\mu_1+\mu_2/\cosh^2 {a}$. The trapping can occur
389: at the only
390: existing minimum $X_1$ (see curve $1$ in the inset of
391: Figure \ref{fig1}) as shown
392: by the trajectory $1$ of Figure \ref{fig1}. As $a$ increases,
393: the potential
394: barriers separate and a local minimum $X_3$ appears, as shown by curves
395: $2-4$ in the inset of Figure \ref{fig1}. If a new minimum appears,
396: the trapping can
397: occur also at the second microshort, as shown by the trajectory $2$
398: of Figure \ref{fig1}. Plots of the potential $U(X)$
399: clearly demonstrate that the shape of the barrier will be optimal
400: when the minimum at $X=X_3$ is quite shallow. Since the half-width
401: of the function $\cosh^{-2}(X)$ is of the order of unity,
402: it is expected that optimal separation of barriers occurs at $a \sim 2$.
403: Numerical evaluation of $a_r$ confirms this estimate:
404: $a_r=1.94$ (for $\mu_1=0.4$, $\mu_2=0.6$); $a_r=2.24$
405: ($\mu_1=0.3$, $\mu_2=0.7$) and $a_r=2.36$ (for $\mu_1=0.2$, $\mu_2=0.8$).
406: If $\mu_1\ge\mu_2$, the transmission scenario is always
407: determined by the first microshort and the trapping occurs only at
408: $X=X_1$. Therefore the dependence $\gamma_{thr}$ on $a$ is
409: monotonically decreasing as shown in Figure \ref{fig2}
410: for $\mu_1=0.6$, $\mu_2=0.4$.
411:
412: It would be of interest to compare how the threshold pinning current
413: depends on the dissipation parameter $\alpha$ and the ratio of the
414: microshort amplitudes $\mu_1$ and $\mu_2$. Since the resonant
415: distance $a_r$ weakly depends on $\mu_{1,2}$ and
416: the pinning current depends strongly on the dissipation constant, it
417: is convenient to normalize $\gamma_{thr}(a,\mu_1,\mu_2;2)$ to the
418: pinning current on the strongest microshort \\
419: $\max{[\gamma_{thr}(\mu_1;1),\gamma_{thr}(\mu_2;1)]}$.
420: In Figure \ref{fig3} the dependence of the enhancement factor
421: %-----------------------------------------------------------------------
422: \begin{equation}
423: \eta(a,\mu_1,\mu_2;2)=
424: \frac{\gamma_{thr}(a,\mu_1,\mu_2;2)}
425: {\max{[\gamma_{thr}(\mu_1;1),\gamma_{thr}(\mu_2;1)]}}~,
426: \label{9}
427: \end{equation}
428: %-----------------------------------------------------------------------
429: on the ratio $\mu_1/(\mu_1+\mu_2)$ for different values of
430: dissipation is shown. The value of the distance between the microshorts
431: has been fixed to $a=2$. Increase of dissipation does not change
432: much the resonant values of $\mu_{1,2}$.
433: However, the value of the enhancement factor at the minimum
434: decreases significantly.
435: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
436: %
437: %-------------------Figure 3-------------------------------------------
438: \begin{figure}[htb]
439: %\begin{center}
440: \centerline{\epsfig{file=fig3.eps,width=2.8in,angle=-90}}
441: \caption{Dependence of the enhancement factor (\ref{9}) on the
442: ratio $\mu_1/(\mu_1+\mu_2)$ $\alpha=0.1$ ($\circ$), $\alpha=0.2$ ($\diamond$)
443: and $\alpha=0.3$ ($+$). Solid lines are used as a guide for an eye.
444: Results, obtained via direct numerical integration of the
445: SG equation with the 4th order Runge-Kutta method
446: for $\alpha=0.3$ are shown by $\oplus$. The inset shows trajectories in the
447: phase plane $(X,v)$ for $\alpha=0.1$, $\gamma=0.06$ (black line);
448: $\alpha=0.2$, $\gamma=0.12$ (red line) and $\alpha=0.3$, $\gamma=0.18$
449: (blue line). Other parameters are $\mu_1=0.4$, $\mu_2=0.6$, $a=2$.}
450: \label{fig3}
451: \end{figure}
452: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
453: In the inset of Figure \ref{fig3} comparison of the fluxon slowing
454: down on the microshorts is shown for different values of dissipation
455: and dc bias. Note that the ratio $\alpha/\gamma$ was kept constant in
456: order to fix the equilibrium velocity $v_\infty$. For stronger
457: dissipation the
458: fluxon slows down to smaller velocities (compare the black and
459: blue curves that correspond to $\alpha=0.1$
460: and $\alpha=0.3$, respectively). Therefore after release from
461: the microshort the fluxon can accelerate to greater values of velocity.
462: As a result, it has more kinetic energy to pass the second microshort.
463: In other words, for larger dissipation one
464: needs larger bias, $\gamma$. Therefore, the tilt of the
465: potential $U(X)$ increases and it smears out the
466: inhomogeneities, created by the impurities.
467:
468: It should be emphasized
469: that the validity of the perturbation theory approach has
470: been confirmed
471: by the direct numerical integration of the original perturbed
472: SG equation (\ref{1}).
473: In Figure \ref{fig2}, $\gamma_{thr}$ has been computed via integration of
474: equation (\ref{1}) for $\mu_1=0.4$ and $\mu_2=0.6$.
475: It is evident that the
476: perturbation theory gives qualitatively the same result and the
477: quantitative difference is not very large. Similarly,
478: in Figure \ref{fig3} the
479: results of the numerical integration
480: of equation (\ref{1}) with $\alpha=0.3$ are
481: given alongside with the perturbation theory results. A good
482: qualitative and
483: quantitative correspondence between these two types of results is
484: clearly demonstrated. Therefore the usage of the approximation
485: (\ref{2})-(\ref{3}) is justified.
486:
487:
488:
489: %######################################################################
490: \subsection{Transmission through $N>2$ microshorts}
491: \label{s3.2}
492: %######################################################################
493:
494: Now we extend the results of the previous subsection on the case of
495: more then $N=2$ microshorts. In Figure \ref{fig4} the dependence of
496: $\gamma_{thr}$ on $a$ for $N=2,3,4,5$ is presented.
497: It clearly demonstrates that addition of an extra impurity to the left
498: from the weakest one decreases further the minimum of the threshold
499: current.
500: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
501: %
502: %-------------------Figure 4-------------------------------------------
503: \begin{figure}[htb]
504: %\begin{center}
505: \centerline{\epsfig{file=fig4.eps,width=2.9in,angle=-90}}
506: \caption{
507: Threshold pinning current as a function of the distance
508: between the microshorts for
509: $\alpha=0.1$, $N=2$, $\mu_1=0.4$, $\mu_2=0.6$ ($+$);
510: $N=3$, $\mu_1=0.27$, $\mu_2=0.4$, $\mu_3=0.6$ ($\nabla$);
511: $N=4$, $\mu_1=0.18$, $\mu_2=0.27$, $\mu_3=0.4$, $\mu_4=0.6$ ($\circ$)
512: and
513: $N=5$, $\mu_1=0.12$, $\mu_2=0.18$, $\mu_3=0.27$, $\mu_4=0.4$, $\mu_5=0.6$
514: ($\times$). Solid lines are used as an guide for an eye.
515: The inset depicts the effective potential $U(X)$ [see
516: equation (\ref{6b})]
517: for the configurations, described in the main figure at $\gamma=0.05$
518: and for the same values of $\{ \mu_n \}$: $N=2$, $a=2$ (curve 1);
519: $N=3$, $a=1.7$ (curve 2); $N=4$, $a=1.6$ (curve 3);
520: $N=5$, $a=1.6$ (curve 4) and $a=1$ (curve 5).
521: }
522: \label{fig4}
523: \end{figure}
524: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
525:
526: The explanation can be easily seen with the help of the effective potential
527: $U(X)$ [see equation (\ref{6b})]. Its shape changes significantly when
528: extra microshorts are added. Comparing curves 1 and 2 in the inset of
529: Figure \ref{fig4} one can see that the energy barrier, which the fluxon
530: should cross, lowers. Adding yet another microshort further lowers
531: the barrier (see curves 3 and 4), so that in the interval $0<X<(N-1)a$
532: the potential barrier almost turns into the decaying slope which is
533: less steep then $-2 \pi \gamma X$. Decrease of $a$ leads to the gradual
534: raising of this slope (compare curves 4 and 5) and consequently to
535: increase of $\gamma_{thr}$.
536:
537:
538: %######################################################################
539: \subsection{Transmission through microresistors}
540: \label{s3.3}
541: %######################################################################
542:
543: A microresistor is an attracting impurity, therefore the fluxon
544: accelerates when approaching it and decelerates back to $v=v_\infty$
545: after passing through or remains trapped if its velocity (and
546: consequently the external bias current) is less than the threshold
547: value. The effective potential $U(X)$ for a microresistor corresponds
548: to the potential well. If two different microresistors are added
549: consequently, the fluxon can be trapped on the first or on the second
550: one, or, if the bias is large enough, pass through.
551: In Figure \ref{fig5} the phase portraits for the system with $N=2$
552: microresistors is shown. The change of the shape of $U(X)$ for
553: the different distances between the microresistors is shown
554: in the inset of Figure \ref{fig5}.
555: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
556: %
557: %-------------------Figure 5-------------------------------------------
558: \begin{figure}[htb]
559: \centerline{\epsfig{file=fig5.eps,width=2.8in,angle=-90}}
560: \caption{
561: Phase space trajectories of equations (\ref{2})-(\ref{3})
562: for $N=2$ microresistors
563: with $\mu_1=-0.7$, $\mu_2=-0.3$ and distance $a=2.5$ between them.
564: Curve $1$ corresponds to $\gamma=0.015$, curve $2$ - to $\gamma=0.03$
565: and curve $3$ - to $\gamma=0.045$. Dissipation for all cases is
566: $\alpha=0.1$. The inset shows the effective potential
567: $U(X)$ [equation (\ref{6b})]
568: for $\mu_1=-0.7,\mu_2=-0.3$, $\gamma=0.03$, $a=1$ (curve 1),
569: $a=2.5$ (curve 2), $a=3.5$ (curve 3) and $a=5$ (curve 4).
570: Red vertical bars denote locations of the microresistors and the lengths
571: of the bars is proportional to their amplitudes, $\mu_{1,2}$.
572: }
573: \label{fig5}
574: \end{figure}
575: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
576:
577: The computation of the threshold current $\gamma_{thr}$ shows that
578: resonant fluxon transmission is possible if $\mu_1 < \mu_2$ and
579: does not happen if $\mu_1 \ge \mu_2$ (see Figure \ref{fig6}a).
580: Explanation of this phenomenon is similar to the case of two
581: microshorts. If the microresistors are located very close to each other,
582: then their amplitudes add up and the fluxon interacts with the
583: microresistor of the amplitude $\mu \simeq \mu_1+\mu_2$. When the
584: impurities start to separate, the effective energy barrier which
585: the fluxon should surmount, lowers (compare curve $1$ with curves $2$
586: and $3$ in the inset Figure \ref{fig6}). The distance
587: between the wells becomes optimal for the best fluxon transmission
588: before they are completely separated (compare curves $3$ and $4$).
589:
590: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
591: %
592: %-------------------Figure 6-------------------------------------------
593: \begin{figure}[htb]
594: \centerline{\epsfig{file=fig6.eps,width=2.7in,angle=-90}}
595: \caption{
596: Panel (a).
597: Threshold pinning current as a function of the distance
598: between two microresistors for $\mu_1=-0.3$, $\mu_2=-0.7$ ($\Delta$);
599: $\mu_1=\mu_2=-0.5$ ($\circ$);
600: $\mu_1=-0.6$, $\mu_2=-0.4$ ($+$); $\mu_1=-0.7$, $\mu_2=-0.3$ ($\times$);
601: $\mu_1=-0.8$, $\mu_2=-0.2$ ($\nabla$); $\mu_1=-0.9$,
602: $\mu_2=-0.1$ ($\diamond$)
603: and $\mu_1=-0.97$, $\mu_2=-0.03$ ($\oplus$). Damping coefficient
604: equals $\alpha=0.1$.
605: Panel (b). Enhancement factor $\eta$ [see equation (\ref{9})] as
606: a function of $a$
607: for $\mu_1=-0.8$, $\mu_2=-0.2$ and different values of damping $\alpha$:
608: $\alpha=0.3$ ($+$), $\alpha=0.2$ ($\times$), $\alpha=0.1$ ($\circ$)
609: and $\alpha=0.05$ ($\nabla$). }
610: \label{fig6}
611: \end{figure}
612: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
613: In contrast to the transmission through two microshorts, the resonant
614: value $a=a_r$ depends strongly on the amplitudes $\mu_{1,2}$. Indeed,
615: for $\mu_1=-0.7$, $\mu_2=-0.3$ one obtains $a_r\simeq 3.62$ and
616: for $\mu_1=-0.9$, $\mu_2=-0.1$ the resonant distance equals
617: $a_r\simeq 2.35$. Comparing curves $2-4$ in the inset of
618: Figure \ref{fig5}, one can notice that the fluxon needs enough kinetic
619: energy to overcome the second maximum, located at $X=X_4$.
620: Obviously, if $X_2$ and $X_4$ are not enough separated, the fluxon will
621: have no time to accelerate in order to avoid trapping on the
622: second microresistor. Therefore the case of curve $3$ is the most
623: optimal one: the height of the barrier at $X_2$ is not too large,
624: as compared to the curve $4$ and the distance between $X_2$ and $X_3$
625: is enough to gain velocity, sufficient for the successful passage
626: over the second barrier. These considerations, of course,
627: correspond to the situation, when the impurities are not
628: strongly separated.
629: Otherwise only the interaction with the first one would matter.
630: For the same reason the position of the minimal threshold current, $a_r$,
631: (see Figure \ref{fig6}b) increases with decrease of the damping parameter
632: $\alpha$. Depth of the minimum
633: decays with decrease of $\alpha$ similarly to the
634: case of two microshorts because with the stronger bias the fluxon
635: can pass through the impurities much easier.
636:
637: Putting additional microresistors after $\mu_2$,
638: $0>\mu_N> \cdots >\mu_2>\mu_1$ further lowers the critical pinning
639: current similarly to the case of $N>2$ microshorts, described
640: in the previous Subsection.
641:
642:
643: %######################################################################
644: \subsection{Transmission through a microshort and a microresistor}
645: \label{s3.4}
646: %######################################################################
647:
648: Finally, we consider the case when two impurities of different polarity
649: (a microshort and a microresistor) are placed one after another.
650: If the microresistor is located before the
651: microshort ($\mu_1<0, \mu_2>0$) resonant enhancement
652: of the threshold pinning current does not happen. In Figure \ref{fig7}
653: (panel a) the phase portraits for this case are shown. The microresistor
654: is an attracting impurity after which the fluxon slows down. On contrary,
655: the microshort is a repelling impurity and the fluxon slows down when
656: approaching it. Therefore it is obvious that by placing impurities in such
657: a way one increases $\gamma_{thr}$ as compared to the case of each
658: individual impurity. The analysis of the effective potential $U(X)$,
659: shown in Figure \ref{fig7} (panel c), further confirms above considerations.
660: The height of the effective barrier, which the fluxon should overcome,
661: can be greater than the height of the individual barriers, created by the
662: individual impurities. If the impurities are very close their
663: influences cancel
664: each other and the fluxon interacts with the impurity of the strength
665: $-|\mu_1|+\mu_2$. The dependence of the threshold pinning
666: current $\gamma_{thr}$
667: on the distance between the impurities is given
668: in Figure \ref{fig8} (panel a).
669: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
670: %
671: %-------------------Figure 7-------------------------------------------
672: \begin{figure}[htb]
673: %\begin{center}
674: \centerline{\epsfig{file=fig7.eps,width=2.5in,angle=-90}}
675: \caption{
676: Phase portraits of the fluxon dynamics and effective potentials
677: for the fluxon dynamics in the case of one microshort and
678: one microresistor.
679: Panel (a) shows fluxon trajectories on the phase plane $(X,v)$
680: for $\alpha=0.1$, $\mu_1=-0.4$, $\mu_2=0.6$, $a=4$; $\gamma=0.05$ (curve 1),
681: $\gamma=0.06$ (curve 2) and $\gamma=0.07$ (curve 3).
682: Panel (b) shows fluxon trajectories on the phase plane $(X,v)$
683: for $\alpha=0.1$, $\mu_1=0.2$, $\mu_2=-0.8$, $a=3$; $\gamma=0.03$ (curve 1),
684: $\gamma=0.0314$ (curve 2) and $\gamma=0.05$ (curve 3).
685: Panel (c) shows the effective potential $U(X)$ for
686: $\mu_1=-0.4$, $\mu_2=0.6$ and different values of $a$:
687: $a=0.1$ (curve 1), $a=1$ (curve 2), $a=2$ (curve 3), $a=3$ (curve 4) and
688: $a=5$ (curve 5).
689: Panel (d) shows the effective potential $U(X)$ for
690: $\mu_1=0.2$, $\mu_2=-0.8$ and different values of $a$:
691: $a=1$ (curve 1), $a=1.7$ (curve 2), $a=2.5$ (curve 3) and $a=4$ (curve 4).
692: }
693: \label{fig7}
694: \end{figure}
695: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
696: For $\mu_1=-\mu_2=-0.5$ the microshort and microresistor cancel each
697: other for $a=0$, therefore the dependence starts at zero and increases
698: until it reaches the maximal value and then decreases, tending
699: monotonically to the value of $\gamma_{thr}$ that corresponds to one
700: microshort with $\mu=0.5$. The dependence of $\gamma_{thr}$ on $a$
701: shows an ``antiresonant'' behaviour because it has a maximum
702: at some certain value of $a$. Analysis of the shape of $U(X)$ from
703: the Figure \ref{fig7}c predicts that the worst transmission would occur
704: when the potential well and the barrier, created by the microresistor
705: and the microshort, respectively, separate
706: from each other far enough to create the highest total
707: barrier (see the curve $3$ of Figure \ref{fig7}c), but not too far (as for the
708: curve $5$ of the same figure) so that each impurity interacts
709: individually with the fluxon.
710:
711: Consider now the case $\mu_1>0, \mu_2<0$. The phase portraits for the
712: fluxon dynamics are shown in Figure \ref{fig7} (panel b). The dependence
713: of the threshold pinning current on the distance between the
714: impurities is shown in Figure \ref{fig8} (panel b). In the case
715: $\mu_1=-\mu_2=0.5$ at $a=0$ the impurities cancel each other. When
716: $a$ increases, $\gamma_{thr}$ monotonically increases, tending to the
717: threshold value of one isolated microshort with the amplitude $\mu=0.5$.
718: In this case trapping occurs only on the microshort because
719: analysis of equations (\ref{4a})-(\ref{4b}) shows that
720: $\gamma_{thr}(0.5;1)>\gamma_{thr}(-0.5;1)$.
721: If $\mu_1 > |\mu_2|$ the dependence of $\gamma_{thr}$ on $a$ is also
722: monotonic. At $a \simeq 0$ the fluxon ``feels'' both impurities as one
723: microshort with $\mu_1-|\mu_2|$. When the impurities separate, the contribution
724: of the microresistor to the total amplitude weakens and the threshold
725: current gradually increases till the value $\gamma_{thr}(\mu_1;1)$.
726: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
727: %
728: %-------------------Figure 8-------------------------------------------
729: \begin{figure}[htb]
730: %\begin{center}
731: \centerline{\epsfig{file=fig8.eps,width=2.5in,angle=-90}}
732: \caption{
733: Threshold pinning current as a function of the distance
734: between two impurities $\mu_1 \mu_2 <0$ with $\alpha=0.1$.
735: Panel (a): $\mu_1=-0.2$, $\mu_2=0.8$ ($\Delta$);
736: $\mu_1=-0.4$, $\mu_2=0.6$ ($\times$); $\mu_1=-0.5$, $\mu_2=0.5$ ($\circ$);
737: $\mu_1=-0.6$, $\mu_2=0.4$ ($+$); $\mu_1=-0.8$, $\mu_2=0.2$ ($\nabla$)
738: and $\mu_1=-0.9$, $\mu_2=0.1$ ($\diamond$).
739: Panel (b): $\mu_1=0.8$, $\mu_2=-0.2$ ($\Delta$); $\mu_1=0.5$, $\mu_2=-0.5$ ($\circ$);
740: $\mu_1=0.4$, $\mu_2=-0.6$ ($+$); $\mu_1=0.3$, $\mu_2=-0.7$ ($\times$);
741: $\mu_1=0.2$, $\mu_2=-0.8$ ($\nabla$) and $\mu_1=0.1$, $\mu_2=-0.9$ ($\diamond$).
742: Dashed lines correspond to the values $\gamma_{thr}(\pm 1; 1)$.
743: Inset shows more detailed picture of $\gamma_{thr}=\gamma_{thr}(a)$.
744: }
745: \label{fig8}
746: \end{figure}
747: %@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@
748: If $\mu_1$ decreases and $|\mu_2|$ increases, behaviour of the
749: critical pinning current on $a$ becomes more complicated. Consider
750: first the case $\mu_1=0.4$, $\mu_2=-0.6$. In the neighbourhood of
751: $a=0$ the system can be considered as a microresistor with the
752: amplitude $\mu_1-|\mu_2|$. When $a$
753: increases, the well and the barrier, created by the microshort $\mu_1$
754: start to separate, increasing the depth of the well (created by the
755: microresistor).
756: After some value of the distance $a$ the dependence of
757: $\gamma_{thr}=\gamma_{thr}(a)$
758: experiences sharp breaking and $\gamma_{thr}$ starts to grow
759: with $a$.
760: The difference between trapping before this breaking point
761: and after it is
762: based on the trapping scenario at $\gamma \le \gamma_{thr}$.
763: For values of $a$ below the breaking point trapping occurs in the well,
764: created by the microresistor
765: (curve $2$ of Figure \ref{fig7}b) while after the breaking
766: point trapping occurs
767: on the microshort. In other words, the breaking point
768: signals the value
769: of the separation of the impurities, before which the fluxon ``feels''
770: them as one microresistor and after which the fluxon ``feels''
771: them separately.
772: Decrease of $\mu_1$, and subsequent increase of $|\mu_2|$
773: leads to the
774: gradual shift of the breaking point to the right and
775: smoothing of the shape of the dependence $\gamma_{thr}(a)$.
776: Further decrease of $\mu_1$ and
777: increase of $|\mu_2|$ makes the dependence $\gamma_{thr}(a)$ more
778: and more flat, so that in the limit $\mu_1 \to 1$, $\mu_2 \to 0$
779: it tends to the horizontal line $\gamma_{thr}=\gamma_{thr}(-1;1)$.
780:
781:
782:
783: %######################################################################
784: %######################################################################
785: \section{Conclusions}
786: \label{s4}
787: %######################################################################
788: %######################################################################
789:
790: We have investigated the fluxon transmission in a dc-biased long Josephson
791: junction (LJJ) through two or more impurities of
792: different polarity: microshorts and/or microresistors. We have observed
793: that the threshold pinning current can depend on the distance between
794: impurities in the resonant way for the case of two or more microshorts
795: or two or more microresistors. That means that at some value
796: of the distance the threshold current attains a minimal value,
797: which is less than the threshold current of the strongest impurity.
798: The resonant transmission does not occur if the fluxon interacts with
799: two impurities of different sign: a microshort and a microresistor.
800:
801: The observed effect should not be confused with the resonant soliton
802: transmission
803: in the non-dissipative cases \cite{ksckv92jpa,zkv94jpsj}.
804: In the case of fluxon dynamics in a long Josephson junction the presence
805: of dissipation is unavoidable. Far away from the impurities fluxon exists
806: as an only one attractor of the system with the velocity, predefined by the
807: damping parameter and external bias. Therefore, contrary to the
808: non-dissipative case, there is no sense in computing the transmission
809: ratio, which in
810: our case can take only two values: zero (trapping) and unity (transmission).
811: Also it should not be confused with the fluxon tunneling
812: as a quantum-mechanical object \cite{sb-jm97prb}
813: across the double-barrier potential, created by two identical
814: microshorts.
815:
816: The discussed phenomenon can be observed experimentally in an annular
817: LJJ via monitoring the current-voltage (IV) characteristics.
818: For a LJJ with one impurity the fluxon IV curve has a hysteresis-like
819: nature with two critical values of the dc bias (discussed in
820: Section \ref{s2}).
821: The lower one is the threshold pinning current, which is the smallest
822: current for which fluxon can propagate. Although simulation in this
823: paper have been performed for the infinite junction, there should be
824: no principal differences with the case of an annular junction with
825: sufficiently large length $L\gg \lambda_J$. Currently experiments are
826: performed in the junctions $L \sim 10\lambda_J$ (see \cite{u98pd}) that
827: can be considered as {\it long}.
828:
829: For the future research in this direction it is of interest to
830: find out how the resonant fluxon transmission changes if the actual
831: size of the impurities and the junction width along the
832: $y$-axis are taken into account.
833:
834: \section*{Acknowledgments}
835: This work has been supported by DFFD, project number GP/F13/088,
836: and the special program of the National Academy of Sciences of Ukraine
837: for young scientists.
838:
839:
840: \begin{thebibliography}{10}
841:
842: \bibitem{barone82}
843: A.~Barone, G.~Paterno,
844: \newblock {\em Physics and Applications of the Josephson Effect},
845: \newblock (Wiley, New York, 1982).
846:
847: \bibitem{u98pd}
848: A.~V. Ustinov, \newblock {Physica D} {\bf 123}, 315 (1998).
849:
850: \bibitem{ms78pra}
851: D.~W. McLaughlin, A.~C. Scott.
852: \newblock {Phys. Rev. A} {\bf 18}, 1652 (1978).
853:
854: \bibitem{restun}
855: A.~F.~J. Levi,
856: \newblock {\em Applied Quantum Mechanics},
857: \newblock (Cambridge University Press, New York, 2006).
858:
859: \bibitem{ksckv92jpa}
860: Yu.~S. Kivshar, A.~Sanchez, O.~Chubykalo, A.~M. Kosevich, L.~Vazquez,
861: \newblock {J. Phys. A: Math. Gen.} {\bf 25}, 5711 (1992).
862:
863: \bibitem{zkv94jpsj}
864: F.~Zhang, Y.~S. Kivshar, L.~Vazquez,
865: \newblock {J. Phys. Soc. Jpn.} {\bf 63}, 466 (1994).
866:
867: \bibitem{parameters} Expressions for the Josephson penetration depth
868: and Josephson plasma frequency and their actual values
869: can be found in \cite{barone82}.
870:
871: \bibitem{kmn88jetp}
872: Yu.~S. Kivshar, B.~A. Malomed, A.~A. Nepomnyashchy,
873: Zh. Eksp. Teor. Fiz. {\bf 94}, 356 (1988)
874: [Sov. Phys. JETP {\bf 67}, 850 (1988)].
875:
876: \bibitem{bp90pla}
877: T.~Bountis, St. Pnevmatikos,
878: \newblock {\em Phys. Lett. A} {\bf 143}, 221 (1990).
879:
880: \bibitem{pb94pla}
881: Th. Pavlopoulos, T.~Bountis.
882: \newblock {\em Phys. Lett. A} {\bf 192}, 215 (1994).
883:
884: \bibitem{sb-jm97prb}
885: A.~Shnirman, E.~Ben-Jacob, B.~Malomed.
886: \newblock {\em Phys. Rev. B} {\bf 56}, 14 677 (1997).
887:
888: \end{thebibliography}
889:
890: \end{document}
891:
892:
893:
894:
895:
896:
897:
898:
899:
900:
901:
902:
903:
904:
905:
906:
907:
908:
909:
910:
911:
912:
913:
914:
915:
916:
917:
918:
919:
920:
921:
922:
923:
924:
925:
926:
927:
928:
929:
930:
931:
932:
933:
934:
935:
936:
937:
938:
939:
940:
941:
942:
943:
944:
945:
946:
947:
948:
949:
950:
951:
952:
953:
954:
955:
956:
957:
958:
959:
960:
961:
962:
963:
964:
965:
966:
967:
968:
969:
970:
971:
972:
973:
974:
975:
976:
977:
978:
979:
980:
981:
982:
983:
984:
985:
986:
987:
988:
989:
990:
991:
992:
993:
994: