1: 

2: \documentstyle[amssymb,preprint,aps]{revtex}

3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4: %TCIDATA{OutputFilter=LATEX.DLL}

5: %TCIDATA{Version=5.00.0.2557}

6: %TCIDATA{<META NAME="SaveForMode" CONTENT="1">}

7: %TCIDATA{Created=Mon Jul 21 11:47:25 2003}

8: %TCIDATA{LastRevised=Thursday, October 13, 2005 19:31:00}

9: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}

10: %TCIDATA{<META NAME="DocumentShell" CONTENT="Journal Articles\REVTeX - APS and AIP Article">}

11: %TCIDATA{Language=American English}

12: %TCIDATA{CSTFile=revtxtci.cst}

13: 

14: \newtheorem{theorem}{Theorem}

15: \newtheorem{acknowledgement}[theorem]{Acknowledgement}

16: \newtheorem{algorithm}[theorem]{Algorithm}

17: \newtheorem{axiom}[theorem]{Axiom}

18: \newtheorem{claim}[theorem]{Claim}

19: \newtheorem{conclusion}[theorem]{Conclusion}

20: \newtheorem{condition}[theorem]{Condition}

21: \newtheorem{conjecture}[theorem]{Conjecture}

22: \newtheorem{corollary}[theorem]{Corollary}

23: \newtheorem{criterion}[theorem]{Criterion}

24: \newtheorem{definition}[theorem]{Definition}

25: \newtheorem{example}[theorem]{Example}

26: \newtheorem{exercise}[theorem]{Exercise}

27: \newtheorem{lemma}[theorem]{Lemma}

28: \newtheorem{notation}[theorem]{Notation}

29: \newtheorem{problem}[theorem]{Problem}

30: \newtheorem{proposition}[theorem]{Proposition}

31: \newtheorem{remark}[theorem]{Remark}

32: \newtheorem{solution}[theorem]{Solution}

33: \newtheorem{summary}[theorem]{Summary}

34: 

35: \begin{document}

36: \title{Reshaping-induced spatiotemporal chaos in driven, damped sine-Gordon

37: systems}

38: \author{R. Chac\'{o}n }

39: \address{Departamento de Electr\'{o}nica e Ingenier\'{\i}a Electromec\'{a}%

40: nica,\\

41: Escuela de Ingenier\'{\i}as Industriales, Universidad de Extremadura, E-06071%

42: \\

43: Badajoz, Spain}

44: \date{\today }

45: \maketitle

46: \pacs{05.45.Gg, 05.45.Yv}

47: 

48: \begin{abstract}

49: Spatiotemporal chaos arising from the competition between

50: sine-Gordon-breather and kink-antikink-pair solitons by reshaping an ac

51: force is demonstrated. After introducing soliton collective coordinates,

52: Melnikov's method is applied to the resulting effective equation of motion

53: to estimate the parameter-space regions of the ac force where homoclinic

54: bifurcations are induced. The analysis reveals that the chaos-order

55: threshold exhibits sensitivity to small changes in the force shape. Computer

56: simulations of the sine-Gordon system show good agreement with these

57: theoretical predictions.

58: \end{abstract}

59: 

60: \section{\protect\bigskip Introduction}

61: 

62: Although nonlinearity is an ubiquitous and fascinating feature of real-world

63: dynamical phenomena, some aspects of such a fundamental property remain

64: poorly explored, partly because they are rarely incorporated into the

65: mathematical models which aim to describe even the simplest of such

66: phenomena. In the context of nonautonomous systems, one of these aspects is

67: the nature of the periodic excitations which can act on a given nonlinear,

68: dissipative or Hamiltonian, system. Since nonlinear systems represent the

69: general case, it would be natural to model periodic excitations (external or

70: parametric) by using periodic functions that are solutions of (physically

71: relevant) nonlinear equations, instead of harmonic functions which are

72: solutions of linear systems, even though the latter have been overwhelmingly

73: used up to now. To consider nonlinear periodic excitations implies enlarging

74: the amplitude-period parameter space to include the parameters, $\alpha _{i}$%

75: , that control the excitation shape. Note that, in the general case, these

76: parameters $\alpha _{i}$ are the (normalized) Fourier coefficients of the

77: periodic function. In physical terms this means that, for a fixed amplitude

78: and period, the parameters $\alpha _{i}$ are responsible for the temporal

79: rate at which energy is transferred from the excitation mechanism to the

80: system. In the context of systems described by ordinary differential

81: equations, this idea has led to the demonstration of the existence of new

82: generic {\it temporal} chaos$\leftrightarrow $order routes by changing only

83: the shape of a nonlinear periodic excitation [1-3]. 

84: 

85: In this present work, the onset of {\it spatio-temporal} chaos associated

86: with the competition between sine-Gordon-breathers and kink-antikink-pairs

87: is investigated in a one-dimensional sine-Gordon (sG) system 

88: \begin{equation}

89: \Phi _{tt}-\Phi _{xx}+\sin \Phi =-\delta \Phi _{t}+\gamma 

90: %TCIMACRO{\func{sn}}%

91: %BeginExpansion

92: \mathop{\rm sn}%

93: %EndExpansion

94: \left[ 4K(m)t/T;m\right] ,  \eqnum{1}

95: \end{equation}%

96: when {\it only} the excitation shape is varied. Here, $%

97: %TCIMACRO{\func{sn}}%

98: %BeginExpansion

99: \mathop{\rm sn}%

100: %EndExpansion

101: \left( u;m\right) $ is the Jacobian elliptic function (JEF) of parameter $m$

102: ($K(m)$ is the complete elliptic integral of the first kind) used {\it solely%

103: } as an example to illustrate the reshaping-induced order-chaos route. When $%

104: m=0$, then $%

105: %TCIMACRO{\func{sn}}%

106: %BeginExpansion

107: \mathop{\rm sn}%

108: %EndExpansion

109: \left[ 4K(m=0)t/T;m=0\right] =\sin (2\pi t/T)$, i.e., one recovers the

110: previously studied case of harmonic excitation [4]. This is relevant in

111: comparing the structural stability of the sG system when solely the

112: excitation shape is varied. In the other limit, $m=1$, one obtains the

113: Fourier expansion of the square wave function: $%

114: %TCIMACRO{\func{sn}}%

115: %BeginExpansion

116: \mathop{\rm sn}%

117: %EndExpansion

118: \left[ 4K(m=1)t/T;m=1\right] =\left( 4/\pi \right) \sum_{n=0}^{\infty

119: }\left( 2n+1\right) ^{-1}\sin \left[ \left( 2n+1\right) 2\pi t/T\right] $.

120: From an experimental standpoint, the choice of the above JEF is motivated

121: for the fact that only the excitation shape is varied by increasing $m$ from

122: 0 to 1, and there is thus a smooth transition from a sine function to a

123: square wave. While these two limiting shapes are readily implemented in

124: experiments, no theoretical analysis connecting them has been developed up

125: to now in the context of sG systems.

126: 

127: The organization of the paper is as follows. In Section 2, a collective

128: coordinate approach [4] is used to obtain an effective equation of motion

129: for the collective variables. In Section 3, Melnikov's method is applied to

130: the resulting effective equation of motion to estimate the parameter-space

131: regions of the elliptic ac force where the onset of chaotic instabilities is

132: expected. Numerical evidence supporting the theoretical predictions from

133: previous sections is presented in Section 4. Finally, Section 5 includes a

134: summary of the findings.

135: 

136: \bigskip 

137: 

138: \bigskip 

139: 

140: \bigskip 

141: 

142: \bigskip 

143: 

144: \section{\protect\bigskip Collective coordinates}

145: 

146: In this present work, a single-breather excitation is assumed to be present

147: in the absence of dissipation and forcing ($\delta =\gamma =0)$. It is well

148: known that the simultaneous action of dissipative $\left( \delta >0\right) $

149: and spatially uniform harmonic forces $\left( \gamma \sin \left( 2\pi

150: t/T\right) \right) $ can break the breather into a kink-antikink pair, which

151: can then recombine into a breather soliton [4,5]. For fixed dissipation,

152: this process may occur repeatedly yielding an intermittent chaotic sequence

153: for certain values of amplitude and period. The above elliptic ac driving

154: permits one to study such a competition between the breather and the

155: kink-antikink pair for parameter values belonging to {\it larger} regions of

156: the amplitude-period parameter space. To this end, one makes a severe mode

157: truncation to the breather collective coordinates [6,7], which yields a set

158: of ordinary differential equations capable of being studied with the aid of

159: Melnikov's method [8,9]. Such an effective equation exhibits a separatrix

160: dynamics which is equivalent to the separatrix dynamics of the full

161: sine-Gordon equation. In particular, homoclinic bifurcations correspond to

162: (chaotic) intermittent binding and unbinding of the kink-antikink pair. In

163: order to determine the equations of motion for the breather collective

164: modes, one first needs the breather and kink-antikink solutions to the

165: unperturbed sine-Gordon equation ($\delta =\gamma =0$ in Eq. (1)) [4].

166: 

167: Breather: 

168: \begin{eqnarray}

169: U_{B}(x,t) &\equiv &4\tan ^{-1}\left\{ \frac{\tan \theta \sin \left[ v\left(

170: t\right) \right] }{\cosh \left[ 2k\left( x-x_{0}\right) \right] }\right\} , 

171: \nonumber \\

172: v(t) &\equiv &v_{0}+t\cos \theta ,  \nonumber \\

173: k &\equiv &\frac{1}{2}\sin \theta ,  \nonumber \\

174: E &\equiv &16\sin \theta .  \eqnum{2}

175: \end{eqnarray}

176: 

177: Kink-antikink pair ($K\stackrel{\_}{K}$): 

178: \begin{eqnarray}

179: U_{K\stackrel{\_}{K}}(x,t) &\equiv &4\tan ^{-1}\left\{ \frac{\sinh \left[

180: u\left( t-t_{0}\right) /(1-u^{2})^{1/2}\right] }{u\cosh \left[ \left(

181: x-x_{0}\right) /(1-u^{2})^{1/2}\right] }\right\} ,  \nonumber \\

182: E_{K\stackrel{\_}{K}} &\equiv &\frac{16}{(1-u^{2})^{1/2}}>16,u\geqslant 0. 

183: \eqnum{3}

184: \end{eqnarray}%

185: Here, $u$ is the $t\rightarrow \infty $ velocity of the kink while $-u$ is

186: the $t\rightarrow \infty $ velocity of the antikink, and $v_{0},x_{0},\theta

187: \left( 0\leqslant \theta \leqslant 2\pi \right) $ are constants. Also, $%

188: \omega _{B}\equiv \cos \theta $ is the breather's internal frequency and $%

189: E,E_{K\stackrel{\_}{K}}$ are the breather and kink-antikink-pair energies,

190: respectively. Following Ref. [4], one assumes the{\it \ ansatz }that the

191: solution of the perturbed sine-Gordon equation has the same form as the

192: unperturbed breather (2), but now allowing $\theta $ and $v_{0}$ to be

193: temporal functions: 

194: \begin{equation}

195: v(t)=v_{0}(t)+\int^{t}\cos \theta \left( t^{\prime }\right) dt^{\prime }. 

196: \eqnum{4}

197: \end{equation}%

198: This approximation means that one is assuming a perturbation sufficiently

199: weak so that its principal effect on the system is to continuously alter the

200: frequency and the phase of the breather. Therefore, such an approximation is

201: valid when $\omega _{B}\ll 1$ since the breather energy is $E_{B}\equiv

202: 16(1-\omega _{B}^{2})^{1/2}$. Next, one can deduce the equations of motion

203: for the breather collective modes $(u,v)$, where 

204: \begin{eqnarray}

205: u(t) &\equiv &U_{B}(x=x_{0},t)=4\tan ^{-1}A(t),  \nonumber \\

206: v(t) &\equiv &\left[ \frac{\partial U_{B}}{\partial t}\right] _{x=x_{0}}=%

207: \frac{4dA(t)/dt}{1+A\left( t\right) ^{2}},  \nonumber \\

208: A(t) &\equiv &\tan \theta \sin v\left( t\right) .  \eqnum{5}

209: \end{eqnarray}%

210: After the coordinate transformation $z\equiv \tan \left( u/4\right) ,%

211: \stackrel{.}{z}\equiv w,$ such equations of motion can be shown to be 

212: \begin{eqnarray}

213: \stackrel{.}{{\bf \chi }} &=&{\bf f}\left( {\bf \chi }\right) +\epsilon {\bf %

214: g}\left( {\bf \chi },\psi \right) ,  \nonumber \\

215: \stackrel{.}{\psi } &=&4K(m)/T,  \eqnum{6}

216: \end{eqnarray}%

217: with 

218: \begin{eqnarray}

219: {\bf \chi } &=&\left( 

220: \begin{array}{c}

221: z \\ 

222: w%

223: \end{array}%

224: \right)   \nonumber \\

225: {\bf f}\left( {\bf \chi }\right)  &=&\left( 

226: \begin{array}{c}

227: w \\ 

228: -z\left( 1-w^{2}\right) \left( 1+z^{2}\right) ^{-1}%

229: \end{array}%

230: \right) ,  \eqnum{7}

231: \end{eqnarray}%

232: and the perturbation 

233: \begin{equation}

234: \epsilon {\bf g}\left( {\bf \chi };t\right) =\left( 

235: \begin{array}{c}

236: 0 \\ 

237: h(t)g_{p}(z,w)-\delta g_{\delta }(z,w)%

238: \end{array}%

239: \right) ,  \eqnum{8}

240: \end{equation}%

241: where 

242: \begin{eqnarray}

243: h(t) &=&\frac{\pi }{4}\gamma 

244: %TCIMACRO{\func{sn}}%

245: %BeginExpansion

246: \mathop{\rm sn}%

247: %EndExpansion

248: \left[ 4K(m)t/T;m\right] ,  \nonumber \\

249: g_{p}(z,w) &=&\frac{\left[ \left( w^{2}+z^{2}\right) z^{2}-s\left\{

250: z^{4}+w^{2}\left[ z^{2}+\left( 1-w^{2}\right) \left( 1+z^{2}\right) \right]

251: \right\} \right] }{\left( 1+sw^{2}\right) \left( w^{2}+z^{2}\right) \left(

252: 1+s\right) \left( 1+z^{2}\right) ^{1/2}}  \nonumber \\

253: &&+\frac{2w^{2}z^{2}s^{2}\left( 1+z^{2}\right) +w^{2}s^{3}\left(

254: 1+2z^{2}\right) \left( 1+z^{2}\right) ^{2}}{\left( 1+sw^{2}\right) \left(

255: w^{2}+z^{2}\right) \left( 1+s\right) \left( 1+z^{2}\right) ^{1/2}}, 

256: \nonumber \\

257: g_{\delta }(z,w) &=&\frac{w\left[ \left( 2+z^{2}\right) \left(

258: w^{2}+z^{2}\right) z^{2}-s\left\{ z^{4}-w^{2}\left[ \left( 2+z^{2}\right)

259: z^{4}-\left( 1-w^{2}\right) \left( 1+z^{2}\right) \right] \right\} \right] }{%

260: \left( 1+sw^{2}\right) \left( w^{2}+z^{2}\right) \left( 1+z^{2}\right) } 

261: \nonumber \\

262: &&+\frac{w\left[ w^{2}s^{2}\left( 1+2z^{2}\right) \left( 1+z^{2}\right) %

263: \right] }{\left( 1+sw^{2}\right) \left( w^{2}+z^{2}\right) \left(

264: 1+z^{2}\right) },  \nonumber \\

265: s &\equiv &\frac{\sinh ^{-1}(z)}{z\left( 1+z^{2}\right) ^{1/2}}.  \eqnum{9}

266: \end{eqnarray}%

267: Noting that the separatrix in the phase-plane $u-v$ is given by 

268: \begin{eqnarray}

269: u_{0,\pm }(t) &=&\pm 4\tan ^{-1}(t),  \nonumber \\

270: v_{0,\pm }(t) &=&\pm 4/(1+t^{2}),  \eqnum{10}

271: \end{eqnarray}%

272: one finds that on the separatrix in the phase-plane $z-w$ 

273: \begin{eqnarray}

274: z_{0,\pm }(t) &=&\pm t,  \eqnum{11} \\

275: w_{0,\pm }(t) &=&\pm 1,  \nonumber

276: \end{eqnarray}%

277: where the positive (negative) sign refers to the upper (lower) homoclinic

278: orbit in the phase-plane $u-v$, which is obtained for the energy $E=16$ of

279: the unperturbed energy functional $E=2\tan ^{2}(u/4)+\left( \stackrel{.}{u}%

280: /4\right) ^{2}\sec ^{2}(u/4).$

281: 

282: \section{\protect\bigskip Chaotic threshold function}

283: 

284: Now one can estimate the threshold of chaotic instabilities for the

285: effective system (6) using Melnikov's method. Thus, the corresponding

286: Melnikov function is 

287: \begin{equation}

288: M^{\pm }(t_{0})=\mp D+C\sum_{n=0}^{\infty }a_{n}(m)p_{n}(T)\sin \left[ \frac{%

289: \left( 2n+1\right) 2\pi t_{0}}{T}\right] ,  \eqnum{12}

290: \end{equation}%

291: where

292: 

293: $\bigskip $%

294: \begin{eqnarray}

295: D &\equiv &\delta \int_{-\infty }^{\infty }\frac{\left( 1-\tanh ^{2}x\right) %

296: \left[ \left( 2-\tanh ^{2}x\right) \tanh ^{2}x+r+r^{2}\left( 1-\tanh

297: ^{4}x\right) \right] }{1+r}dx\simeq 2.03727\delta ,  \nonumber \\

298: p\left( 2\pi /T\right)  &\equiv &\int_{-\infty }^{\infty }\frac{\cos \left( 

299: \frac{2\pi }{T}\sinh x\right) \left[ \left( 1-r\right) \left( 1-\tanh

300: ^{2}x\right) \right] 

301: %TCIMACRO{\func{sech}}%

302: %BeginExpansion

303: \mathop{\rm sech}%

304: %EndExpansion

305: x+2r^{2}\tanh ^{2}x+r^{3}\left( 1+\tanh ^{2}x\right) }{\left( 1+r\right) ^{2}%

306: }dx,  \nonumber \\

307: C &\equiv &\pi ^{2}\gamma /(4\sqrt{m}K(m)),  \nonumber \\

308: a_{n}(m) &\equiv &%

309: %TCIMACRO{\func{csch}}%

310: %BeginExpansion

311: \mathop{\rm csch}%

312: %EndExpansion

313: \left[ \left( 2n+1\right) \pi K(1-m)/(2K(m))\right] ,  \nonumber \\

314: p_{n}(T) &\equiv &p\left[ \left( 2n+1\right) 2\pi /T\right] ,  \eqnum{13}

315: \end{eqnarray}%

316: and where $r\equiv 2x%

317: %TCIMACRO{\func{csch}}%

318: %BeginExpansion

319: \mathop{\rm csch}%

320: %EndExpansion

321: \left( 2x\right) $. From Eq. (12) one sees that a homoclinic bifurcation is

322: guaranteed if 

323: \begin{equation}

324: \frac{\delta }{\gamma }<U\left( m,T\right) ,  \eqnum{14}

325: \end{equation}%

326: where the chaotic threshold function is 

327: \begin{equation}

328: U\left( m,T\right) \equiv \frac{\pi ^{2}}{4\alpha \sqrt{m}K(m)}%

329: \sum_{n=0}^{\infty }\left( -1\right) ^{n}a_{n}\left( m\right) p_{n}(T). 

330: \eqnum{15}

331: \end{equation}%

332: The series in Eq. (15) satisfies the conditions of the Leibnitz theorem on

333: alternate series, and then one can deduce a slightly more restrictive

334: condition for the onset of the aforementioned spatio-temporal chaotic

335: instabilities, in order to facilitate the comparison with the previously

336: studied case [4] of a harmonic force $\left( m=0\right) $: 

337: \begin{equation}

338: \frac{\delta }{\gamma }<U\left( m=0,T\right) R(m),  \eqnum{16}

339: \end{equation}%

340: where $U(m=0,T)=\pi p(2\pi /T)/(4\alpha ),R(m)\equiv \pi \left( \sqrt{m}%

341: K(m)\right) ^{-1}%

342: %TCIMACRO{\func{csch}}%

343: %BeginExpansion

344: \mathop{\rm csch}%

345: %EndExpansion

346: \left[ \pi K(1-m)/(2K(m))\right] $. It is worth noting that the condition

347: (16) is a reliable approximation due to the coefficients $a_{n}(m)\simeq 0$, 

348: $\forall n\geqslant 1$ and $\forall m\in \left[ 0,m^{\ast }\right] $ with $%

349: m^{\ast }$ being very close to 1. A plot of the shape function $R(m)$ is

350: shown in Fig. 1, where one sees that $R(m=0)=1$, as expected. The rapid

351: growth of $R(m)$ as $m\rightarrow 1$ means that the chaotic threshold

352: exhibits hypersensitivity to changes in the excitation shape as this

353: approximates a square wave. This can be understood in terms of the

354: increasing accumulation of {\it effective} harmonics in the Fourier series

355: of the elliptic force as $m$ $\rightarrow $1. Note that this number of

356: effective harmonics is not a linear function of $m$ because of the

357: dependence of the elliptic force on $K(m)$.

358: 

359: \section{Numerical simulations}

360: 

361: For the purpose of the comparison with the above theoretical predictions,

362: the parameters $\delta =0.2,T=2\pi /0.6,$and $L=24$ were fixed, and an sG

363: breather with frequency $\omega _{B}=\sqrt{2}/2$ and $x_{0}=v_{0}=0$ (cf.

364: Eq. (2)) was chosen as initial data. The period $T=2\pi /0.6$ corresponds to

365: a frequency $\omega =0.6$ (for a sinusoidal force, $m=0$) for which the

366: chaotic attractors are dominated by breather to kink-antikink transitions

367: [11,12]. Equation (1) was integrated under periodic boundary conditions $%

368: \Phi \left( x=-L/2,t\right) =\Phi \left( x=L/2,t\right) $ for all $t$.

369: Figure 2 shows an illustrative example of the comparison between the

370: theoretically obtained chaotic threshold amplitude $\gamma _{th}$ (solid

371: line, cf. Eq. (16)) and the corresponding numerically obtained amplitude

372: (dots) as the shape parameter $m$ varies. One sees that the amplitudes

373: obtained numerically are slightly larger than the respective theoretically

374: predicted ones, as expected, although an overall good agreement is found.

375: Similar good agreement was found for other sets of parameters in the

376: large-period regime ($T\gg 2\pi $), provided that the damping and driving

377: force were sufficiently weak. An illustrative sequence of the order-chaos

378: route is shown in Fig. 3 where the initial state ($m=0$) occurs after the

379: system has undergone a spatial period doubling (for $m=0$ and $0.6\lesssim

380: \gamma \lesssim 0.9$, cf. Ref. [12]) but remains simply periodic in time.

381: This two-breather state has thus a spatial period $\frac{1}{2}$ and a

382: temporal period 1, and exhibits mirror symmetry with respect to the origin.

383: The spatial symmetry is broken as the forcing shape deviates from

384: sinusoidal. Figure3 illustrates this phenomenon with the values $m=0.8$ and $%

385: m=0.82$. This spatial symmetry breaking of the two-breather state appears to

386: be an unavoidable precursor of the onset of spatio-temporal chaos via the

387: kink-antikink collision mechanism, as can be appreciated in Fig. 3 for the

388: shape parameter $m=0.898\simeq m_{th}$, which corresponds to the

389: theoretically predicted chaotic threshold (cf. Eq. (16)). For higher values

390: of the shape parameter, i.e., when the driving shape is ever closer to a

391: square wave, one finds the persistence of the chaotic dynamics (cf. Figs. 3, 

392: $m=0.99$) predicted by the Melnikov-method-based analysis.

393: 

394: \section{\protect\bigskip Conclusions}

395: 

396: In sum, it has been shown through the example of an elliptic ac driving that

397: the competition between sine-Gordon-breather and kink-antikink-pair solitons

398: strongly depends on the shape of the spatially uniform ac driving. In

399: particular, it was theoretically predicted and numerically confirmed that a

400: square wave is more effective than a sinusoidal function to yield

401: spatio-temporal chaos associated with the competition between

402: sine-Gordon-breathers and kink-antikink-pairs. Potential application of the

403: present results to reshaping-induced control (suppression and enhancement)

404: of chaotic dynamics [1,14,15] range from DNA molecules [16] to long

405: Josephson junctions and quasi-one-dimensional ferromagnets [17]. An

406: interesting problem under current investigation is whether the above

407: predictions on the chaotic threshold remain reliable in the so-called (for $%

408: m=0$) nonlinear (cubic) Schr\"{o}dinger regime (i.e., $T\lesssim 2\pi $),

409: where the chaotic dynamics is dominated by breather-breather transitions

410: instead of the breather to kink-antikink transitions of the regime studied

411: here ($T\gg 2\pi $), since both mechanisms involve homoclinic crossings [13].

412: 

413: \begin{acknowledgement}

414: The author acknowledges financial support from the Spanish MCyT and the

415: European Regional Development Fund (FEDER) program through project

416: FIS2004-02475.

417: \end{acknowledgement}

418: 

419: \begin{references}

420: \bibitem{1} Chac\'{o}n R, D\'{\i}az Bejarano J. Routes to suppressing chaos

421: by weak periodic perturbations. Phys Rev Lett 1993; 71: 3103-3106.

422: 

423: \bibitem{2} Zeni AR, Gallas JAC. Lyapunov exponents for a Duffing

424: oscillator. Physica D 1995; 89: 71-82.

425: 

426: \bibitem{3} Chac\'{o}n R, Mart\'{\i}nez Garc\'{\i}a-Hoz A. Route to chaos

427: via strange non-chaotic attractors by reshaping periodic excitations.

428: Europhys Lett 2002; 57: 7-13.

429: 

430: \bibitem{4} Ariyasu JC, Bishop AR. Determination of the onset of

431: spatiotemporal complexity induced by soliton competition in a perturbed

432: sine-Gordon chain. Phys Rev A 1989; 39: 6409-6415.

433: 

434: \bibitem{5} Bishop AR, McLaughlin DW, Salerno M. Global coordinates for the

435: breather-kink (antikink) sine-Gordon phase space: An explicit separatrix as

436: a possible source of chaos. Phys Rev A 1989; 40: 6463-6469.

437: 

438: \bibitem{6} Ercolani NM, Forest MG, McLaughlin DW, Montgomery R. Hamiltonian

439: structure of the modulation equations for sine-Gordon wavetrains. Duke Math

440: J 1987; 55: 949-983.

441: 

442: \bibitem{7} Salerno M. Reduced sine-Gordon breather-(anti)kink dynamics and

443: the double sine-Gordon system. Phys Lett A 1989; 134: 421-423.

444: 

445: \bibitem{8} Melnikov VK. On the stability of the center for time periodic

446: perturbations. Trans Moscow Math Soc 1963; 12: 1-57.

447: 

448: \bibitem{9} Guckenheimer J, Holmes PJ. Nonlinear oscillations, dynamical

449: systems, and bifurcations of vector fields. New York: Springer; 1983.

450: 

451: \bibitem{11} Bishop AR, Fesser K, Lomdahl PS, Kerr WC, Williams MB,

452: Trullinger SE. Phys Rev Lett 1986; 50: 1095-1098.

453: 

454: \bibitem{12} Bishop AR, Lomdahl PS. Nonlinear dynamics in driven, damped

455: sine-Gordon systems. Physica D 1986; 18: 54-66.

456: 

457: \bibitem{13} Bishop AR, Forest MG, McLaughlin DW, Overman II EA. A

458: quasi-periodic route to chaos in a near-integrable pde. Physica D 1986; 23:

459: 293-328.

460: 

461: \bibitem{14} R\"{o}delsperger F, Kivshar YS, Benner H. Reshaping-induced

462: chaos suppression. Phys Rev E 1995; 51: 869-872.

463: 

464: \bibitem{15} Tereshko V, Shchekinova E. Resonant control of the R\"{o}ssler

465: system. Phys Rev E 1998; 58: 423-426.

466: 

467: \bibitem{16} Yakushevich LV. Nonlinear physics of DNA. Chichester: Wiley \&

468: Sons; 1998.

469: 

470: \bibitem{17} Kivshar YS, Malomed BA. Dynamics of solitons in near integrable

471: systems. Rev Mod Phys 1989; 61: 763-915.

472: \end{references}

473: 

474: \bigskip 

475: 

476: \subsubsection{Figure Captions}

477: 

478: \bigskip 

479: 

480: Figure 1. Shape function $R(m)$ (see the text) versus the shape parameter $m$%

481: .

482: 

483: \bigskip 

484: 

485: Figure 2. Theoretical (solid line, cf. Eq. (16)) and numerically obtained

486: (dots) chaotic threshold amplitude, $\gamma _{th}$, versus shape parameter, $%

487: m$, for $\delta =0.2$ and $T=2\pi /0.6$.

488: 

489: \bigskip 

490: 

491: Figure 3. Space-time evolution of $\Phi \left( x,t\right) $ for the sG

492: system through four driving periods for $\delta =0.2,\gamma =0.79,T=2\pi /0.6

493: $, and shape parameters $m=0$, $0.5$, $0.8$, $0.82$, $0.898\simeq m_{th}$,

494: and $0.99$.

495: 

496: \end{document}

497: