1: \documentstyle[aps]{revtex}
2: %\documentstyle[aps,preprint]{revtex}
3: \textheight 23.5cm
4: %\headsep 2.cm
5: \topmargin -2cm
6: \textwidth 16.5cm
7: \oddsidemargin 0.46cm
8: \evensidemargin 0.46cm
9: %\parskip=0.1 true cm
10: %\pagestyle{empty}
11: %\def\theequation{\arabic{section}.\arabic{equation}}
12: \newcommand{\ds}{\displaystyle}
13: \newcommand{\dsf}{\ds\frac}
14: \newcommand{\re}[1]{(\ref{#1})}
15: \newcommand{\beq}{\begin{equation}}
16: \newcommand{\eeq}{\end{equation}}
17: \renewcommand{\baselinestretch}{1.2}
18: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
19: \sloppy
20: \large
21: %\draft
22: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23:
24: \begin{document}
25: \large
26:
27: \begin{center}
28: \Large\bf
29:
30: Nonlinear Thermomagnetic Waves in the Resistive State of Superconductors
31: \vskip 0.1cm
32: {\normalsize\bf N.€.\,Taylanov}\\
33: \vskip 0.1cm
34: {\large\em Research Institute of Applied Physics,\\
35: National University of Uzbekistan,\\
36: Vuzgorodok, Tashkent, 700174 Uzbekistan,\\
37: e-mail:taylanov@iaph.tkt.uz}
38: \end{center}
39: \begin{center}
40: \bf Abstract
41: \end{center}
42:
43: \begin{center}
44: \mbox{\parbox{14cm}{\small
45: The evolution of thermomagnetic perturbations in the resistive state
46: of superconductors is considered. A qualitative pattern of the formation and
47: further development of nonlinear stationary structures that describe the final
48: stage of thermal and electromagnetic perturbations in a superconductor is
49: investigated. The wave propagation velocity and the wave front width in a
50: superconductor are estimated.
51: }}
52: \end{center}
53: \vskip 0.5cm
54:
55: Energy dissipation during the motion of vortices leads to ohmic
56: heating of a superconductor. As a result, a certain region of the
57: superconductor is heated to a temperature $T>T_{c}$, where $T_c$ is the
58: critical temperature. An increase in the temperature in a local region
59: of the sample brings about a decrease in the critical current $j_{c}$ and
60: the emergence of vortex electric field $E$ at the same region.
61: The effect of superconductivity quenching due to thermal heating
62: of the vortex lattice has been experimentally studied for a long time. Early
63: experiments [I] revealed that the vortex electric field E is induced under
64: ohmic heating in a superconductor through which a direct current with a
65: density $j_c$ flows. According to the balance between dissipative and
66: nonlinear effects, the transition to the resistive state is accompanied
67: by the emergence of various modes of the "switching wave" type, i.e.,
68: the regime of a wave motion that switches a sample from the superconducting
69: to the normal state. Examples of these modes can be provided by thermal
70: waves, namely, a steady-state propagation of a normal zone [2] or nonlinear
71: thermomagnetic waves [3] in superconductors.
72:
73: In this work, we studied the qualitative pattern of formation and the
74: profile of nonlinear dissipative structures, i.e., stationary traveling
75: waves which describe the final stage of the evolution of thermal and
76: electromagnetic perturbations in the resistive state of superconductors.
77:
78: The evolution of thermal and electromagnetic perturbations in a
79: superconductor is described by the nonlinear one-dimensional heat
80: conduction equation
81:
82: \beq
83: \nu\dsf{dT}{dt}=\kappa\dsf{d^2T}{dx^2}+\vec j\vec E
84: \eeq
85: the Maxwell equation
86: \beq
87: \dsf{d^2\vec E}{dx^2}=\dsf{4\pi}{c^2}\dsf{d\vec j}{dt}\,,
88: \eeq
89: and their related equation of the critical state
90: \beq
91: \vec j=\vec j_{á}(T,\vec H)+\vec j_{r}(\vec E)\,.
92: \eeq
93: where $\nu$ is the heat capacity and $\kappa$ is the thermal conductivity
94: coefficients; $\vec j_{c}$ and $\vec j_{r}$ are the densites of the critical
95: and resistive currents, respectively.
96:
97: The model under consideration is essentially nonlinear, because the
98: right-hand side of Eq. (1) contains a term describing the Joule heat
99: generation in the region of the resistive phase. An exact solution to
100: the essentially nonlinear parabolic partial differential equations
101: (1)-(3) does not exist.
102:
103: Note that the evolution of perturbations of the temperature $T(x,t)$
104: and the fields $E(x,t)$ and $H(x,t)$ is determined by the equation of
105: critical state (3).
106: Because of considerable analytical difficulties,
107: we will restrict ourselves to considering the Bean model [4] and
108: assume that the critical current density is independent of the external
109: magnetic field; i.e., $\dsf{dj_{c}}{dH}=0$. The dependence $j_c(T)$ is
110: described by the relationship $j_{c}(T)=j_{0}-a(T-T_0)$; where $T_ 0$
111: is the initial temperature of the superconductor and the quantity
112: $a=\left|\dsf{dj_c}{dT}\right|_{T=T_0}$
113: describes the thermally activated weakening of the Abrikosov vortex
114: pinning by lattice defects. The dependence $j_r(E)$ in the region of the
115: electric field $E\ge E_f$ ($E_f$ is the boundary of the linear section in the
116: current-voltage characteristic of the superconductor) can be approximated
117: by a piecewise-linear function $j_r(E)\sim \sigma_{f} E$, where $\sigma_f$ is
118: the effective conductivity. The dependence $j_r(E)$ is essentially nonlinear
119: in the flux creep region $E\le E_f$ [5]. Here, we will consider a perturbation
120: with a sufficiently high amplitude and use the linear dependence
121: $j_{r}(E)$.
122:
123: We seek a solution to the initial system of equations as a function
124: of the new self-simulated variable $\xi(x,t)$, that is
125: \beq
126: \begin{array}{ll}
127: T=\Theta[\xi(x,t)]\,,
128: E=E[\xi(x,t)]\,,
129: H=H[\xi(x,t)]\,,
130: \end{array}
131: \eeq
132: which $\xi=x-vt$ describes a traveling wave that moves with a constant velocity
133: $v$ along the $x$ axis [3].
134:
135: After substitution of relationships (4) into the initial system et of equations
136: and performing a trivial differentation, we obtain the following system of equations
137: for the variable $\xi(x,t)$
138:
139: \beq
140: - \nu v\dsf{dT}{d\xi}=\dsf{d}{d\xi}\left[\kappa\dsf{dT}{d\xi}\right]+jE,
141: \eeq
142: \beq
143: \dsf{dE}{d\xi}=-\dsf{4\pi v}{c^2}j,
144: \eeq
145: \beq
146: E=\dsf{v}{c}H.
147: \eeq
148:
149: The corresponding thermal and electrodynamic boundary conditions for
150: Eqs. (5)-(7) have the form
151:
152: \beq
153: \begin{array}{ll}
154: T(\xi\rightarrow+\infty)=T_0\,,&\quad \dsf{dT}{dx}(\xi\rightarrow-\infty)=0\,,\\
155: E(\xi\rightarrow+\infty)=0\,,&\quad E(\xi\rightarrow-\infty)=E_e\,.
156: \end{array}
157: \eeq
158: where $E_e$ is the amplitude of the electric field.
159: It should be noted that inclusion of the temperature dependences of the
160: parameters $\kappa$ and $\nu$ substantially complicates analytical
161: calculations of the wave evolution dynamics that is described by the
162: system of Eqs. (5)- (7). In most cases, the changes in the local values of these
163: parameters in the sample can be considered negligible comparatively to the
164: characteristic scale of temperature variations. Hence, we can take
165: these parameters to be constant.
166: Indeed, the investigation revealed that the thermal
167: conductivity almost does not affect the character of the stationary
168: wave propagation. This stems from the fact that the thermal flux
169: $\kappa\dsf{dT}{d\xi}$ vanishes at stationary points of the system
170: at $\xi\rightarrow\pm \infty$. However, the
171: temperature dependence of the heat capacity should be taken into account.
172: Such a dependence is representeq as
173: $\nu\approx\nu_0\left(\dsf{T}{T_0}\right)$ over a wide range of
174: temperatures [5].
175:
176: By eliminating variables $T(x,t)$ and $H(x,t)$ with the aid of relationships
177: (5) and (7) and employing the boundary conditions (8), we obtain a
178: differential equation for the $E$ wave distribution:
179:
180: \beq
181: \frac{d^2 E}{dz^2}-2\pi\frac{\nu T_0}{E_\kappa}\frac{v^2}{c^2}\left[\left(1+
182: \frac{\sigma_f E}{aT_0}+\frac{c^2}{4\pi avT_0}\frac{dE}{dz}\right)^{4}-1
183: \right]+
184: \beta\tau\frac{dE}{dz}=\frac{E^2}{2E_\kappa}\,.
185: \eeq
186: Here, we introduced the following dimensionless parameters:
187: $$
188: z=\dsf{\xi}{L}\,,\quad
189: L=\frac{cH_e}{4\pi j_0}\,,\quad
190: E_\kappa=\frac{\kappa}{aL^2}\,,\quad
191: \beta=\frac{vt_\kappa}{L}\,,\quad
192: \tau=\frac{4\pi\sigma_f\kappa}{c^2\nu}\,,\quad
193: t_\kappa=\frac{\nu L^2}{\kappa}\,,
194: $$
195: where $L$ is the depth of the magnetic field penetration into the sample
196: and $t_\kappa$ is the thermal time of the problem.
197: According to the qualitative theory [6], the equilibrium states are
198: found from the condition
199: \beq
200: 2\pi\nu_0 T_0\frac{v^2}{c^2}
201: \left[\left(1+\frac{\sigma_f E}{aT_0}\right)^{4}-1\right]=E^2\,.
202: \eeq
203:
204: An evident property of system ( 10) is the absence of closed curves that
205: are fully composed of the phase trajectories in the phase plane
206: $\left(E,\dsf{dE}{d\xi}\right)$.
207: The proof of this statement can be based on the Bendixson criterion [7] .
208: The number of stationary points ( one or three) and their
209: type are determined by the parameter
210: \beq
211: P=2\pi\nu_0T_0\frac{v^2}{c^2}\,.
212: \eeq
213:
214: The three equilibrium points $E=0$, $E=E_1$ and $E=E_2$ correspond to the
215: condition $P<P_k=\dsf{1}{2}$ (Fig.1). There is only one singular point
216: $E_0=0$ at $P>P_k$. The parabola and the quartic curve in Eq. (10) are
217: tangent at $P=P_k$; i.e., this condition corresponds to the coincidence
218: $E_1=E_2=E^*=\dsf{6}{7}\frac{aT_0}{\sigma_{f}}$.
219: The direct solution of Eq. (10) yields the following waves:
220: \begin{eqnarray}
221: \begin{array}{lll}
222: 1)& E_{1,2}=E^*[1+2,2(P_k-P)^{1/2}]\,,&\quad\mbox{¯à¨}\quad
223: \left(\dsf{P_k-P}{P_k}\right)<<1\,;\\
224: \quad\\
225: 2)& E_1=8\pi\dsf{\sigma_f\nu_0}{a} \frac{v^2}{c^2}\,,&\\
226: \quad\\
227: & E_2=(2\pi)^{1/2} \dsf{\sigma_f^2\nu_0^2}{a^2}\frac{c}{v}>>E_1\,,&
228: \quad\mbox{at}\quad P_k>>P\,.
229: \end{array}
230: \end{eqnarray}
231: Analysis of the phase plane shows that the points $E_0=0$ and $E=E_2$
232: are stable nodes and that $E=E_1$ is a saddle. In addition to the separatrix
233: $E_1 E_0$, the set (10) has the separatrix $E_1 E_2$ connecting the points
234: $E_1$ and $E_2E_1$ (Fig.2). This means that two types of waves with amplitudes
235: $\Delta E=E_1$ and $\Delta E=E_2-E_1$ can exist in the superconductor.
236: Evidently, wave 1) has an amplitude of the order $E_k$ at $P\rightarrow P_k$;
237: its velocity is determined by equality (10) at
238: $E=E_{1}$. Equation (10) has two stationary points at $P<<P_k$: $E_{0}=0$
239: is a stable node and $E_{1}=2\beta^2\tau E_{\kappa}$ is a saddle.
240:
241:
242: The separatrix that connects these two equilibrium states corresponds to a
243: "difference"-type solution with amplitude $E_{e}$, which is related to the wave
244: velocity $v_{E}$ and the wave front width $\Delta z$ by the following
245: equations:
246:
247: \beq
248: v_{E}=\frac{L}{t_{\kappa}}\left[\dsf{E_{e}}{2\tau E_{\kappa}}\right]^{1/2}\,,
249: \eeq
250: \beq
251: \Delta z=16\dsf{1+\tau}{\tau^{1/2}}\left[\dsf{E_{\kappa}}{E_{e}}\right]^{1/2}\,.
252: \eeq
253: Wave 2) has a small amplitude at $P\rightarrow P_k$
254: \beq
255: \frac{\Delta E}{E_k}=4,4(P_k-P)^{1/2} <<1\,,
256: \eeq
257: and its velocity is inversely proportional to the amplitude at $P<<P_k$.
258: Such an exotic dependence of $v_E$ on $\Delta E =E_e$ most likely means that
259: the waves of this type are unstable. Note that observation of the
260: second-type waves becomes possible in finite-sized samples with asymmetric
261: boundary conditions.
262: In conclusion, it should be noted that the above investigations prove the
263: possibility of applying the results obtained to high-temperature
264: superconductors cooled to liquid-nitrogen temperatures (T = 77 K), providing
265: that the values of the physical parameters of the sample are known.
266:
267: \begin{center}
268: {\bf REFERENCE}
269: \end{center}
270: \begin{enumerate}
271: \item
272: V.A.Al'tov, V.B.Zenkevich, M.G.Kremlev, and V.V.Sychev,
273: Stabilization of Superconducting Magnetic Systems (Energoatomizdat, Moscow,
274: 1984).
275: \item
276: A.V.Gurevich and R.G.Mints, eat Autowaves in Normal Metals and
277: Superconductors (Nauka, Moscow, 1987).
278: \item
279: N.A.Taylanov Superconduct. Science and Technology, 14, 326 (2001).
280: \item
281: C. P. Bean, Phys. Rev. Lett. 8 (6),250 (1962).
282: \item
283: R. G. Mints and A. L. Rakhmanov, Instabilities in Superconductors
284: (Nauka, Moscow, 1984).
285: \item
286: V. I. Karpman, Nonlinear Waves in Dispersive Media
287: (Nauka, Moscow, 1973; Pergarnon, Oxford, 1975).
288: \item
289: A.A.Andronov, A.A.Vitt, S.E.Khaykin, Theory of Oscillators
290: (Nauka, Moscow, 1981).
291: \end{enumerate}
292:
293:
294:
295: \end{document}
296:
297:
298:
299:
300:
301: