nlin0701032/bkm2.tex
1: \documentclass[floats,aps,showpacs,twocolumn]{revtex4}
2: 
3: \usepackage[dvips]{graphicx}
4: \usepackage{amsmath}   % for \text
5: \usepackage{amsfonts}
6: 
7: 
8: 
9: \renewcommand{\topfraction}         {1.0}
10: \renewcommand{\bottomfraction}      {1.0}
11: \renewcommand{\textfraction}        {0.0}
12: \renewcommand{\floatpagefraction}   {0.0}
13: 
14: \newcommand{\heff}{h_{\text{eff}}}
15: \newcommand{\heffbar}{\hbar_{\text{eff}}}
16: \newcommand{\Areg}{A_{\text{reg}}}
17: \newcommand{\Dreg}{\Delta_{\text{reg}}}
18: \newcommand{\Dcha}{\Delta_{\text{ch}}}
19: \newcommand{\Ncha}{N_{\text{ch}}}
20: \newcommand{\Nreg}{N_{\text{reg}}}
21: \newcommand{\Ntot}{N_{\text{tot}}}
22: \newcommand{\Wch}{W_{\text{ch}}}
23: \newcommand{\freg}{f_{\text{reg}}}
24: \newcommand{\cFreg}{{\cal F}_{\text{reg}}}
25: \newcommand{\ffl}{f_{\text{{f}l}}}
26: \newcommand{\tH}{\tau_{\text{\tiny{H,ch}}}}
27: 
28: \newcommand{\mmax}{m_{\text{max}}}
29: \newcommand{\mstar}{m^{*}}
30: \newcommand{\mfl}{\mstar}
31: 
32: %\newcommand{\PSImagx}[2]{\includegraphics[width=#2]{psplots/#1}}
33: \newcommand{\PSImagx}[2]{\includegraphics[width=#2]{#1}}
34: 
35: \newcommand{\Z}{\mathbb{Z}}
36: \newcommand{\N}{\mathbb{N}}
37: 
38: \newcommand{\ud}{\text{d}}
39: \newcommand{\ue}{\text{e}}
40: \newcommand{\ui}{\text{i}}
41: 
42: 
43: \newcommand{\emm}{m}  
44: \newcommand{\Wnull}{W_0}
45: \newcommand{\Weins}{W_1}
46: \newcommand{\Wm}{W_\emm}
47: \newcommand{\Weq}{W_{\text{eq}}}
48: 
49: 
50: \begin{document}
51: 
52: \title{Universality in the flooding of regular islands by chaotic states}
53: 
54: \author{Arnd B\"acker$^1$, Roland Ketzmerick$^1$, 
55: and Alejandro G. Monastra$^{1,2}$}
56: 
57: \affiliation{$^1$Institut f\"ur Theoretische Physik, Technische
58: Universit\"at Dresden, 01062 Dresden, Germany\\
59: $^2$Departamento de F\'{\i}sica, Comisi\'on Nacional de Energ\'{\i}a At\'omica, 
60: Av. del Libertador 8250, 1429 Buenos Aires, Argentina}
61: 
62: \date{05.01.2007, revised 03.05.2007}
63: 
64: \begin{abstract}
65: 
66: We investigate the structure of eigenstates in 
67: systems with a mixed phase space
68: in terms of their projection onto individual regular tori.
69: Depending on dynamical tunneling rates and the Heisenberg time,
70: regular states disappear and chaotic states flood the regular tori.
71: For a quantitative understanding
72: we introduce a random matrix model.
73: The resulting statistical properties of
74: eigenstates as a function of an effective coupling strength
75: are in very good agreement with numerical results for a kicked system.
76: We discuss the implications of these results for the applicability
77: of the semiclassical eigenfunction hypothesis.
78: \end{abstract}
79: \pacs{05.45.Mt, 03.65.Sq}
80: 
81: \maketitle
82: 
83: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
84: \section{Introduction}
85: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
86: 
87: 
88: The classical dynamics in Hamiltonian systems shows a rich behaviour
89: ranging from integrable to fully chaotic motion. In chaotic systems
90: nearby trajectories separate exponentially in time and ergodicity
91: implies that a typical trajectory fills out the energy-surface in a
92: uniform way. However, integrable and fully chaotic dynamics are
93: exceptional \cite{MarMey74} as typical Hamiltonian systems show a
94: mixed phase space in which regions of regular motion, the
95: so-called regular islands around stable periodic orbits, and
96: chaotic dynamics, the so-called chaotic sea, coexist.
97: 
98: For quantized Hamiltonian systems the fundamental questions concern
99: the behaviour of the eigenvalues and the properties of eigenfunctions,
100: especially in the semiclassical regime.
101: From the semiclassical eigenfunction hypothesis
102: \cite{Per73,Vor76,Ber77b,Vor79,Ber83} one expects that in the
103: semiclassical limit the eigenstates concentrate on those regions in
104: phase space which a typical orbit explores in the long-time limit. For
105: integrable systems these are the invariant tori.
106: In contrast, for ergodic systems almost all orbits fill
107: the energy shell in a uniform way. For this situation
108: the semiclassical eigenfunction hypothesis 
109: is proven by the quantum ergodicity theorem which shows
110: that almost all eigenstates become equidistributed on the energy 
111: shell~\cite{Qerg}.
112: 
113: For systems with a mixed phase space, in the semiclassical limit $(h
114: \rightarrow 0)$, the semiclassical eigenfunction hypothesis implies
115: that the eigenstates can be classified as being either regular or
116: chaotic according to the phase-space region on which they
117: concentrate. This is supported by several studies, see
118: e.g.~\cite{BohTomUll93,ProRob93b,LiRob95b,CarVerFen98,VebRobLiu99,MarKee2005}.
119: It is also possible, that the influence of a
120: regular island  quantum mechanically extends beyond the outermost
121: invariant curve due to partial barriers like cantori 
122: and that quantization conditions remain 
123: approximately applicable even outside of the island \cite{BohTomUll93}.
124: However, it was recently shown that 
125: the classification into regular and chaotic states
126: does not hold when the phase
127: space has an infinite volume \cite{HufKetOttSch2002}. In this case
128: eigenstates may completely ignore the classical phase space boundaries
129: between regular and chaotic regions.
130: 
131: 
132: In order to understand the behaviour of eigenstates away from the
133: semiclassical limit, i.e.\ at finite values of the Planck constant
134: $h$, one has to compare the size of phase-space structures with $h$.
135: Let us consider for simplicity the case of two-dimensional area
136: preserving maps and their quantizations. Regular states of an island
137: concentrate on tori which fulfill the EBK-type
138: quantization condition
139: \begin{equation} \label{eq:EBK}
140:  \oint p \, \ud q = (m+1/2) h \qquad m=0, 1, ...
141: \end{equation}
142: for the enclosed area \cite{BerBalTabVor79}. This quantization rule
143: explicitly shows that regular eigenstates only appear if $h/2$ is
144: smaller than the area $\Areg$ of that island.
145: 
146: Another consequence of finite $h$ in systems with a mixed phase space
147: is dynamical tunneling
148: \cite{DavHel81}, i.e.\ tunneling through dynamically generated
149: barriers in phase space, in contrast to the usual tunneling under a
150: potential barrier. Dynamical tunneling couples the subspace spanned by
151: the regular basis states, corresponding to the quantization condition
152: \eqref{eq:EBK}, with the complementary subspace \cite{ftn2}
153: composed of chaotic
154: basis states. This raises the question whether the {\it eigenstates}
155: of such a quantum system
156: can still be called {\it regular} or {\it chaotic}.
157: 
158: 
159: 
160: In Ref.~\cite{BaeKetMon2005} it was shown that \eqref{eq:EBK} is not a
161: sufficient condition for the existence of a regular eigenstate on the
162: $m$-th quantized torus. In addition one has to fulfill
163: \begin{equation}
164:   \gamma_m < \frac{1}{\tH} , \label{newcondition}
165: \end{equation}
166: where $\tH = h / \Dcha$ is the Heisenberg time of the surrounding
167: chaotic sea with mean level spacing $\Dcha$ and $\gamma_m$ is the
168: decay rate of the $m$-th regular state, if the chaotic sea were
169: infinite.
170: When condition \eqref{newcondition} is violated one observes
171: eigenstates which extend over the chaotic region and flood the $m$-th
172: torus \cite{BaeKetMon2005}. To distinguish them from the chaotic
173: eigenstates that do not flood the torus, they are referred to as 
174: {\it flooding eigenstates}. For the limiting case of complete flooding of
175: all tori, the corresponding eigenstates were called amphibious
176: \cite{HufKetOttSch2002}. Recently, the consequences of flooding for the
177: transport properties in rough nano-wires were 
178: studied \cite{FeiBaeKetRotHucBur2006}.
179: 
180: 
181: 
182: The process of flooding was explained and
183: demonstrated for a kicked system 
184: in Ref.~\cite{BaeKetMon2005}.
185: Condition \eqref{newcondition} was
186: obtained by scaling arguments, which cannot provide a
187: prefactor. Moreover, for an ensemble of systems, one would like to know
188: the probability for the existence of a regular
189: eigenstate. In particular, when varying the Heisenberg time, how broad
190: is the transition regime during which this probability goes from 1 to 0? 
191: Another question is, how do the chaotic eigenstates turn into flooding
192: eigenstates for a given torus?
193: 
194: In this paper we give quantitative answers to these questions. We
195: study the flooding of regular tori in terms of the weight of
196: eigenstates inside the regular region and devise a random matrix model
197: which allows for describing the statistics of these weights in
198: detail. Random matrix models have been very successful for obtaining
199: quantitative predictions on eigenstates in both fully chaotic systems
200: and systems with a mixed phase space, see
201: e.g.~\cite{KusMosHaa1988,BohTomUll93,HaaZyc1990,TomUll1994,Bae2003,KeaMezMon2003}.
202: For the present situation we propose a random matrix model which takes
203: regular basis states and their coupling to the chaotic basis states
204: into account. The only free parameters are the strength of the
205: coupling and the ratio of the number of regular to the number
206: of chaotic basis states. From this
207: model the weight distribution for eigenstates is determined.
208: 
209: 
210: For a kicked system we define the weight by the projection of the
211: eigenstates onto regular basis states localized on a given torus $m$. The
212: distribution of the weights allows for studying the flooding of each torus
213: separately. The resulting distributions are compared with the
214: prediction of the random matrix model and, after an appropriate
215: rescaling, very good agreement is observed. This agreement shows
216: explicitly the universal features underlying the process of flooding,
217: giving a precise criterion for the existence or non--existence of
218: regular, chaotic, and flooding eigenstates in mixed systems.
219: 
220: The text is organized as follows. In section \ref{TheSystem} we
221: introduce the kicked system used for the numerical illustrations, both
222: classically (part A) and quantum mechanically (part B). In section
223: \ref{TheSystem}~C we define the weight of an eigenstate by its projection
224: onto regular basis states and investigate the distribution of the
225: weights for the kicked system. In section \ref{RMModel} we 
226: introduce the random matrix model
227: and determine the corresponding weight distribution as a function of
228: the coupling strength. In section \ref{Comparison} the relation
229: between parameters of the kicked system and the random matrix model is
230: derived. This allows for a direct comparison of the distributions. In
231: section \ref{Fraction} we consider the fraction of regular
232: eigenstates, both for an individual torus and for the entire island. 
233: In section \ref{Fraction-flood} we briefly discuss  
234: the consequences of the random matrix model on the 
235: number of flooding eigenstates. 
236: A summary and discussion of the eigenfunction
237: structure in generic systems with a mixed phase space
238: is given in section \ref{Conclusions}.
239: 
240: 
241: 
242: 
243: 
244: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
245: \section{The Kicked System} \label{TheSystem}
246: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
247: 
248: %=======================================================================
249: \subsection{Classical dynamics}
250: %=======================================================================
251: 
252: 
253: 
254: \begin{figure}[b]
255:   \begin{center}
256:     \PSImagx{three_cells_class.eps}{8.6cm}
257:     %\PSImagx{three_cells_class_small.eps}{8.6cm}
258:     \caption{(color online) 
259:               (a) Classical dynamics of the kicked system given by 
260:               Eqs.~(\ref{mapQ}) and (\ref{mapP}). 
261:               Invariant tori of the regular island are
262:               shown (continuous curves) and the transport to the right is
263:               indicated by the arrows. The dots correspond to one
264:               chaotic orbit. The magnification shows that the
265:               boundary of the island to the chaotic sea is rather
266:               sharp with only very small secondary islands.
267:               (b) Smoothed functions $T'(p)$ and $V'(q)$ (blue, dark lines)
268:               and discontinuous functions $t'(p)$ and $v'(q)$ (red, light
269:               lines) according to Eqs.~(\ref{tp}-\ref{Vq}).
270:      \label{fig:system}} 
271:    \end{center}
272: \end{figure}
273: 
274: 
275: For a general one-dimensional kicked Hamiltonian
276: \begin{equation}
277:   H(p,q,t) =  T(p) + V(q) \sum_{n= -\infty}^{\infty} \delta(t - n ) \ ,
278:    \label{hamiltonian}
279: \end{equation}
280: the dynamics is fully determined by the mapping of position and
281: momentum $(q_n,p_n)$ at times $t=n + 0^+$ just after the kicks
282: \begin{eqnarray}
283:   q_{n+1} &=& q_n + T'(p_n) \ , \label{mapQ} \\
284:   p_{n+1} &=& p_n - V'(q_{n+1}) \ . \label{mapP}
285: \label{mapping}
286: \end{eqnarray}
287: Choosing the functions $T'(p)$ and $V'(q)$ appropriately, one can
288: obtain a system with a large regular island and a homogeneous chaotic
289: sea.  For the system considered in \cite{BaeKetMon2005}, first
290: introduced in \cite{HufKetOttSch2002}, one starts with
291: the piecewise linear functions (see Fig.~\ref{fig:system}b)
292: \begin{eqnarray}
293:   t'(p) &=& \frac{1}{2} + \left( \frac{1}{2} - s p \right) {\rm sign}
294:   \left( p - \lfloor p + 1/2 \rfloor \right) \ , \label{tp} \\
295:   v'(q) &=& -r q - (1-r) \lfloor q + 1/2 \rfloor \ , \label{vq}
296: \end{eqnarray}
297: where $\lfloor x \rfloor$ is the floor function, and $s$ and $r$ are
298: two parameters determining the properties of the regular island
299: and the chaotic sea. 
300: Using a Gaussian smoothing with $G_{\varepsilon}
301: (z)=\exp(-z^2/2\varepsilon^2)/ \sqrt{2\pi\varepsilon^2}$, one obtains
302: analytic functions
303: \begin{eqnarray}\label{functionsprime}
304:   T'(p) &=& \int_{-\infty}^{\infty} \ud z \; t'(p+z) \; G_{\varepsilon}(z) \ ,
305:             \label{Tp} \\
306:   V'(q) &=& \int_{-\infty}^{\infty} \ud z \; v'(q+z) \; G_{\varepsilon}(z)  \ .
307:             \label{Vq}
308: \end{eqnarray}
309: By construction, these functions have the periodicity properties
310: \begin{eqnarray}\label{periodicityprime}
311:   T'(p+k) &=& T'(p)   \ , \\
312:   V'(q+k) &=& V'(q)-k  \ ,\label{periodicityprime2}
313: \end{eqnarray}
314: for any integer $k$. We consider $p\in[-1/2, 1/2[$ and
315: $q\in[-1/2,-1/2+M[\,$ with periodic boundary conditions. The phase space
316: is composed of a chain of transporting islands centered at $(\bar
317: q,\bar p)=(k,1/4)$ with $0 \leq k \leq M-1$ that are mapped one unit
318: cell to the right (see Fig.~\ref{fig:system}a). The surrounding chaotic
319: sea has an average drift to the left as the overall transport is
320: zero \cite{SchOttKetDit2001,SchDitKet2005}. 
321: The fine scale structure at the boundary of the island to the
322: chaotic sea has a very small area (see the magnification in
323: Fig.~\ref{fig:system}a). Resonances in this layer are irrelevant in
324: the $h$ regime studied here.
325: For $s=2$, $r=0.65$
326: and $\varepsilon=0.015$ the regular island has a relative area $\Areg
327: \approx 0.215$.
328: 
329: 
330: 
331: %=======================================================================
332: \subsection{Quantization} \label{Quantization}
333: %=======================================================================
334: 
335: 
336: In kicked systems, the quantum evolution of a state after one period of time
337: \begin{equation}
338:   |\psi (t + 1) \rangle = \hat{U} |\psi (t) \rangle \ , \label{evolution}
339: \end{equation}
340: is fully determined by the unitary operator, see e.g.\
341: \cite{BerBalTabVor79,HanBer80,ChaShi86,Esp93,DegGra03},
342: \begin{equation}
343:   \hat{U}= \exp\left( -\frac{2 \pi \ui}{\heff} V( \hat{q} ) \right) 
344:            \exp\left( -\frac{2 \pi \ui}{\heff} T( \hat{p} ) \right) 
345:                              \ . \label{propagator}
346: \end{equation}
347: Here the effective Planck's constant $\heff$ is Planck's constant $h$
348: divided by the size of one unit cell. The eigenstates of this operator
349: are defined by
350: \begin{equation} \label{eq:eigenstates}
351:   \hat{U} |\psi_j \rangle = \ue^{2 \pi \ui \varphi_j} |\psi_j \rangle  \ , 
352: \end{equation}
353: where the eigenphase $\varphi_j$ is the quasienergy divided by 
354: $\hbar\omega$. 
355: In order to fulfill the periodicity of the classical dynamics
356: in $p$ direction, the quantum states have to obey the
357: quasi-periodicity condition
358: \begin{equation}
359:   \langle p + 1 |\psi \rangle 
360:     = \ue^{- 2\pi\ui \chi_p} \langle p  |\psi \rangle  \ .
361:   \label{quantumperiodP}
362: \end{equation}
363: One can show that this leads to quantum states that are a linear
364: combination of the discretized position states $| q_j \rangle$, with
365: $q_j = \heff ( j + \chi_p)$. Additionally, imposing periodicity after
366: $M$ unit cells in $q$ direction, quantum states have to fulfill the
367: property
368: \begin{equation}
369:   \langle q + M |\psi \rangle = 
370:      \ue^{2\pi\ui \chi_q} \langle q  |\psi \rangle  \ . \label{quantumperiodQ}
371: \end{equation}
372: Because of the required periodicity the phase space is compact and the
373: effective Planck's constant can only be a rational number
374: \begin{equation}
375:   \heff = \frac{M}{N}   \ . \label{quanthbar}
376: \end{equation}
377: We consider the case of incommensurate $M$ and $N$, so that the
378: quantum system is not effectively reduced to less than $M$ cells.
379: 
380: 
381: 
382: \begin{figure}[t]
383:   \begin{center}
384:     \PSImagx{param_var.eps}{8.6cm}
385:     %\PSImagx{param_var_small.eps}{8.6cm}
386: 
387:     \caption{(color online) (a) Eigenphases of the kicked system vs
388:               $\chi_q$ for $\heff=1/10$. The pattern of straight lines
389:               (interrupted by avoided crossings) with negative slope
390:               corresponds to regular eigenstates with $m=0$ and $m=1$
391:               whose Husimi functions are shown to the right. The other
392:               eigenstates are chaotic and live outside of the
393:               regular region, as can be seen from the Husimi
394:               representation. (b) Weights $\Wnull$ and $\Weins$ of all
395:               eigenstates vs $\chi_q$ (left). Distribution $P(W)$ of
396:               these weights in a log-linear representation (right).
397:               }
398:               \label{fig:Avoided} 
399:   \end{center}
400: \end{figure}
401: 
402: The properties \eqref{periodicityprime}, \eqref{periodicityprime2}
403: of $T'(p)$ and $V'(q)$ imply for their integrals
404: \begin{eqnarray}
405: T(p+k) &=& T(p)   \ , \label{periodicityT} \\
406: V(q+k) &=& V(q)- k q - \frac{k^2}{2} \ . \label{periodicityV}
407: \end{eqnarray}
408: From this one finds that the propagator $\hat U$ is consistent with
409: the periodicity conditions \eqref{quantumperiodP} and
410: \eqref{quantumperiodQ} if and only if
411: \begin{equation}
412:   M \left( \chi_p + \frac{N}{2} \right)  \in \mathbb{Z}  \ . 
413:   \label{quantcondition}
414: \end{equation}
415: For given $M$ and $N$, this condition limits the possible values of
416: the phase $\chi_p$, while $\chi_q$ remains arbitrary. Thus, in the
417: basis given by the position states $|q_j \rangle$, with $0 \leq j \leq
418: N-1$, where $N$ is the dimension of the Hilbert space, the propagator
419: $\hat U$ is represented by the finite $N \times N$ unitary matrix
420: 
421: \begin{equation} \label{Umatrix}
422:    U_{k l} = \frac{1}{N} \sum_{j=0}^{N-1} \ue^{ -\frac{\ui}{\heffbar} 
423:            \left[ V(q_k) + T(p_j) + p_j (q_l - q_k) \right]}    \ , 
424: \end{equation}
425: where $0 \leq k,l \leq N-1$ and $p_j = (j+\chi_q)/N$. Finding the
426: solution of \eqref{eq:eigenstates}, i.e. the eigenphases and
427: eigenstates of the system, therefore reduces to the numerical
428: diagonalization of the matrix \eqref{Umatrix}.
429: The result is illustrated in
430: Fig.~\ref{fig:Avoided}(a) for $\heff = 1/10$, where the eigenphases
431: are plotted as a function of $\chi_q$. The straight lines with
432: negative slope correspond to the regular eigenstates
433: \cite{SchOttKetDit2001,SchDitKet2005}, whose Husimi distributions are
434: shown to the right in Fig.~\ref{fig:Avoided}(a).  Lines with an
435: average positive slope correspond to chaotic eigenstates.
436: 
437: When the system consists of $M$ unit cells one has $M$ regular basis
438: states localized on the $m$-th torus. Their EBK eigenphases are
439: equispaced with a distance $1/M$ \cite{Sch:PhD-BaeKetMonSch}.
440: 
441: 
442: 
443: %=======================================================================
444: \subsection{Projection onto regular basis states} \label{Projection}
445: %=======================================================================
446: 
447: 
448: 
449: In order to investigate the amount of flooding we use the projection
450: of the eigenstates onto regular basis states of the island region.
451: For the considered kicked system regular basis states can be
452: constructed from harmonic oscillator eigenstates, as the invariant
453: tori are accurately approximated by ellipses
454: \cite{Sch:PhD-BaeKetMonSch}. The expression for
455: the $m$-th harmonic oscillator state, centered in a phase space point
456: $(\bar q,\bar p)$, is
457: \begin{align}
458:    \langle q &| \varphi^{\emm}_{\bar q,\bar p} \rangle =
459:    \frac{1}{\sqrt{2^m m!}} \left( \tfrac{{\rm Re}~\sigma}{\pi \heffbar}
460:    \right)^{1/4} H_m \left( \sqrt{\tfrac{{\rm Re}~\sigma}{\heffbar}} (q
461:    - \bar q) \right) \nonumber\\
462:     & \times \exp \left( - \tfrac{\sigma}{2 \heffbar} (q- \bar q)^2 +
463:     \tfrac{\ui}{\heffbar} \bar p (q - \bar q /2) \right) 
464:    \label{tiltedstate}
465: \end{align}
466: where $H_m$ is the Hermite polynomial of degree $m$. The complex
467: constant $\sigma$ takes into account the squeezing and rotation of the
468: state. From the linearized map at the stable fixed point of the island
469: one finds $\sigma = (\sqrt{351}- 13\, \ui )/40$.
470: 
471: 
472: For a chain with $M$ identical 
473: cells, a regular basis state is a linear combination of
474: the harmonic oscillator states $|\varphi^{\emm}_{k,1/4} \rangle$,
475: centered in the $k$-th island for $0 \leq k \leq M-1$ and properly
476: normalized and periodized in the $q$ and $p$ directions
477: \cite{Sch:PhD-BaeKetMonSch}. The subspace spanned by these $M$ regular basis
478: states is the same as the one spanned by the $M$ harmonic oscillator
479: states $|\varphi^{\emm}_{k,1/4} \rangle$.  Therefore, we define the
480: weight $\Wm$ of a normalized state $| \Psi \rangle$ by its projection
481: onto this subspace corresponding to the $m$-th quantized torus
482: \begin{equation}
483:   \Wm = \sum_{k=0}^{M-1} | \langle \Psi | \varphi^{\emm}_{k,1/4}
484:   \rangle |^2 \ . \label{weightsemiclass}
485: \end{equation}
486: 
487: \begin{figure}[b]
488:   \begin{center}
489:     \PSImagx{fig_m0.eps}{8.6cm}
490:     \caption{(color online) 
491:               Distribution of $\Wnull$ (Eq.~\eqref{weightsemiclass})
492:               vs system size $M$ for effective Planck's constant
493:               $\heff \approx 1/10$.
494:     \label{fig:Weights_m0} } 
495:   \end{center}
496: \end{figure}
497: 
498: \begin{figure}[t]
499:   \begin{center}
500:     \PSImagx{fig_m1}{8.6cm}
501:     \caption{(color online) 
502:              Distribution of $\Weins$ (Eq.~\eqref{weightsemiclass})
503:               vs system size $M$ for effective Planck's constant
504:               $\heff \approx 1/10$.
505:     \label{fig:Weights_m1} } 
506:   \end{center}
507: \end{figure}
508: 
509: 
510: By means of the weight $\Wm$ for all eigenstates of
511: Eq.~\eqref{Umatrix} we can study the process of flooding for each
512: torus separately. This allows for a detailed analysis and a
513: quantitative comparison with a random matrix model. Therefore this is
514: a considerable improvement compared to our previous analysis
515: \cite{BaeKetMon2005}, where the weight was defined as the integral of
516: the Husimi distribution of an eigenstate over the whole region
517: of the island, which means that the information on individual 
518: tori is not accessible.
519: 
520: In Fig.~\ref{fig:Avoided}(b) we show the weights $\Wnull$ and $\Weins$
521: of all the eigenstates as a function of $\chi_q$. For $\Wnull$ we
522: observe that for almost all $\chi_q$ the weights are essentially zero
523: or one. Only at avoided crossings of regular and chaotic eigenstates
524: their weights have intermediate values. For $m=1$ the avoided
525: crossings are much broader due to the larger coupling and the value
526: $W=1$ is not reached between several avoided crossings. This is also
527: seen in the weight distributions shown to the right in
528: Fig.~\ref{fig:Avoided}(b), where the two peaks from the chaotic
529: eigenstates (at $W=1$) and from the regular eigenstates (at $W=0$) are
530: broader for $m=1$ in comparison with $m=0$. Note, that in the situation
531: of isolated avoided crossings the involved eigenstates are often
532: referred to as hybrid states.
533: 
534: 
535: 
536: The distribution of the weights $\Wm$ allows for studying the process
537: of flooding in a quantitative way. To violate condition
538: \eqref{newcondition} we need to increase the Heisenberg time, while keeping
539: the tunneling rates $\gamma_m$ constant. We can achieve this by
540: choosing a sequence of rational approximants $M/N$ of $\heff = 1 / (d
541: + g)$, with $d\in\N$ and the golden mean $g = (\sqrt{5} - 1)/2 \approx
542: 0.618$. This ensures that, while the system size $M$ is increased, 
543: $\heff$ is essentially kept at a fixed
544: value, and therefore the tunneling rates $\gamma_m$ are independent of
545: $M$. Simultaneously, the dimensionless Heisenberg time $\tH = 1 /
546: \Dcha$ increases linearly with $M$,
547: \begin{equation} \label{eq:tH}
548:   \tH = \Ncha = \left( \frac{1}{\heff} - \mmax \right) M  \ ,
549: \end{equation}
550: where we used $\Dcha = 1 / \Ncha$ and $\Ncha = N - \mmax M$
551: is the number of chaotic states. 
552: Here $\mmax$ is the maximal number of regular states 
553: in a single island according to the EBK quantization condition
554: \eqref{eq:EBK}, 
555: $\mmax = \left\lfloor \Areg / \heff + 1/2 \right\rfloor$.
556: As discussed in Ref.~\cite{BaeKetMon2005},
557: $\tH$ may be bounded, due to 
558: localization effects: For $M$ larger than the
559: localization length $\lambda$ the effective mean level spacing $\Dcha
560: \sim (\lambda \Ncha/M)^{-1}$ leads to 
561: $\tH \sim \lambda \Ncha/M \approx \lambda \heff$,
562: where $\lambda$ is measured in multiples of a unit cell and
563: $\Ncha/M$ is the number of chaotic states per unit cell.
564: For transporting islands, like in
565: the model studied here, $\lambda \sim 1/\gamma_0$ is unusually large
566: \cite{HanOttAnt1984,IomFisZas2002andrefs,HufKetOttSch2002},
567: leading to a maximal value $\tH \sim \heff/\gamma_0$.
568: 
569: In Figs.~\ref{fig:Weights_m0} and \ref{fig:Weights_m1} we show the
570: distribution of $\Wnull$ and $\Weins$ for $d=9$ (giving approximants
571: $\heff = 1/10$, $2/19$, $3/29$, $5/48$, $\ldots$) 
572: for increasing system size
573: $M$. For small system sizes we increased the statistics by varying the
574: phase $\chi_q$ in the quantization, as it was shown in
575: Fig.~\ref{fig:Avoided}(b). To present the results in a compact form
576: each histogram is shown using a color scale. The horizontal strips
577: for $M=1$ in Fig.~\ref{fig:Weights_m0} and Fig.~\ref{fig:Weights_m1}
578: correspond to the histograms previously shown in Fig.~\ref{fig:Avoided}(b).
579: 
580: In Fig.~\ref{fig:Weights_m0} one clearly observes for small $M$
581: two separate peaks corresponding to chaotic eigenstates 
582: at $W=0$ \cite{ftn}
583: and regular eigenstates with $m=0$ at $W=1$. With increasing system
584: size these regular eigenstates disappear while the weight $\Wnull$ of
585: the chaotic eigenstates starts to increase 
586: and they turn into flooding eigenstates.
587: 
588: 
589: 
590: Comparing Fig.~\ref{fig:Weights_m1} for $\Weins$ with
591: Fig.~\ref{fig:Weights_m0} for $\Wnull$ one observes a qualitatively
592: similar behavior. The difference is that the regular eigenstates with
593: $m=1$ disappear for much smaller system size $M \approx 100$ than the
594: eigenstates with $m=0$, as expected from Eq.~\eqref{newcondition} and their
595: ratio of tunneling rates, $\gamma_0 / \gamma_1 \ll 1$.
596: 
597: For the largest values of $M$ only flooding eigenstates are left which
598: fully extend over the chaotic sea and the regular island. The flooding
599: is complete and the $N$ eigenstates are equally distributed in the
600: Hilbert space. Projecting them onto the $M$ regular basis states leads
601: to the average value $\Weq = M/N = \heff \approx 1/10$, in agreement
602: with the observed position of the peaks in Figs.~\ref{fig:Weights_m0}
603: and \ref{fig:Weights_m1} and the findings in
604: Ref.~\cite{HufKetOttSch2002}.
605: 
606: 
607: 
608: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
609: \section{Random Matrix Model} \label{RMModel}
610: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
611: 
612: 
613: 
614: The similarity in the behavior of the histograms in
615: Fig.~\ref{fig:Weights_m0} and Fig.~\ref{fig:Weights_m1}, 
616: suggests a universality in the
617: process of flooding, which should allow for a random matrix modelling.
618: Such models have been used  successfully for the case of mixed systems
619: to describe the level splitting in the context of chaos assisted 
620: tunneling, see e.g.~\cite{BohTomUll93,TomUll1994,LeyUll96,ZakDelBuc98}. 
621: %
622: In our case, we want to describe the statistics of 
623: eigenvectors for the situation of a chain of $\Nreg$ regular islands.
624: Here one has $\Nreg$ equispaced
625: regular levels corresponding to the $m$-th quantized torus and $\Ncha$
626: COE distributed chaotic levels coupled by dynamical tunneling, see
627: Fig.~\ref{fig:ladder}. For this situation we propose a random matrix
628: model with the following block structure
629: %
630: \newcommand{\ENTRY}[1]{\raisebox{0ex}[5mm][3mm]{\makebox[7ex]{$#1$}}}
631: \begin{equation}  \label{matrix}
632: H =
633: \left(
634: \begin{array}{c|c}
635: \ENTRY{H_{\rm reg}} & \ENTRY{V}           \\[0.2ex]\hline
636: \ENTRY{V^T}         & \ENTRY{H_{\rm ch}}
637: \end{array}
638: \right) \ .
639: \end{equation}
640: This matrix is chosen to be real symmetric because the kicked system 
641: under consideration obeys time reversal symmetry.
642: As a consequence of the block structure, the free parameters
643: of this model are the ratio $\Nreg/\Ncha$ of the
644: number of regular and chaotic basis states and the strength $v$ 
645: of the coupling.
646: 
647: \begin{figure}[t]
648:   \begin{center}
649:     \PSImagx{ladder.eps}{8.6cm}
650:     \caption{Schematical plot of the regular levels with spacing $\Dreg$
651:              coupled with strength $v\Dcha$ to the COE distributed
652:              chaotic levels with mean spacing $\Dcha$.
653:     \label{fig:ladder} } 
654:   \end{center}
655: \end{figure}
656: 
657: The first block $H_{\rm reg}$ models the regular basis states
658: associated with one specific torus, while for simplicity we neglect
659: the regular basis states quantized on other tori. As discussed at the
660: end of Sec.~\ref{TheSystem}~B, in the considered kicked system, the
661: EBK eigenphases of the $\Nreg$  regular basis
662: states are equispaced. To mimic this behavior we consider for $H_{\rm
663: reg}$ a diagonal matrix with elements $(k+\chi)/\Nreg$, $k=0, 1,
664: \ldots, \Nreg -1$. The parameter $\chi$ can be chosen from a uniform
665: distribution between zero and one. The energies lie in the interval [0,1] with
666: fixed spacing $\Dreg = 1 / \Nreg$.
667: 
668: \begin{figure}[b]
669:   \begin{center}
670:     \PSImagx{fig_new_RMT_model.eps}{8.6cm}
671:     \caption{(color online) Distribution of weights in the random
672:               matrix model vs coupling strength $v$. The ratio
673:               $\Nreg/\Ncha$ approximates the value $1/(8+g)$.  The
674:               dashed lines at $W=0.5$ and $W = 0.5 \Weq
675:               \approx 0.052$ separate 
676:               chaotic, flooding, and regular eigenstates.
677:     \label{fig:Weights_RMT} } 
678:   \end{center}
679: \end{figure}
680: 
681: The block $H_{\rm ch}$ models the $\Ncha$ chaotic basis states, where
682: we assume $\Ncha>\Nreg$. It is also a diagonal matrix whose elements
683: $\{ E_l \}$ are the eigenphases of an $\Ncha \times \Ncha$ matrix of
684: the Circular Orthogonal Ensemble (COE). These energies $\{ E_l \}$ lie
685: in the interval [0,1] with a uniform average density and show the
686: typical level repulsion of chaotic systems. The mean level spacing of
687: these basis states is $\Dcha = 1 / \Ncha$. Note, that a GOE matrix for
688: this block would have been less convenient as it leads to a
689: non-uniform density of levels according to Wigner's semicircle law.
690: 
691: 
692: The off-diagonal block $V$ accounts for the coupling between the
693: regular and chaotic basis states. It is a $\Nreg \times \Ncha$
694: rectangular matrix, where each element is a random Gaussian variable
695: with zero mean and variance $(v\Dcha)^2$. The positive parameter $v$
696: is the coupling strength in units of the chaotic mean level spacing
697: $\Dcha$. Thereby the results become asymptotically independent of the
698: dimension $\Ntot = \Nreg + \Ncha$ of the matrix for fixed $v$ and
699: $\Nreg/\Ncha$.
700: 
701: 
702: We identify the regular region with the subspace spanned by the first
703: $\Nreg$ components. Therefore, for any normalized vector $(\Psi_0,
704: \ldots, \Psi_{\Nreg-1}, \Psi_{\Nreg}, \ldots, \Psi_{\Ntot-1})$ 
705: we define the weight $W$ inside the regular region as
706: \begin{equation}
707: W = \sum_{j=0}^{\Nreg-1} | \Psi_j |^2 \ . \label{weightrmm}
708: \end{equation}
709: For a particular realization of the ensemble through the numbers
710: $\{E_l \}$, $\chi$, and the block $V$, we compute the weights $W$ of
711: the eigenvectors. We take for the statistics only those eigenvectors
712: whose eigenenergies are in the interval [0.1, 0.9] to avoid possible
713: border effects.  We determine the distribution of $W$ by averaging
714: over many different realizations. Increasing the matrix size $\Ntot$
715: for a fixed ratio $\Nreg/\Ncha$ we find that the distribution
716: converges. Considering a ratio $\Nreg/\Ncha = 1/(8+g)$ and a small
717: coupling strength $v \approx 0.1$ the distribution converges around
718: $\Ntot = 200$. For $v \approx 1$ bigger matrices of $\Ntot
719: \approx 1000$ are necessary. For $v \approx 10$, we used $\Ntot
720: \approx 10000$. The limiting distributions depend sensitively on the
721: coupling strength $v$.
722: 
723: 
724: In Fig.~\ref{fig:Weights_RMT} we plot the distribution of $W$ for
725: different values of $v$.  We have to distinguish between the uncoupled
726: regular and chaotic basis states of our model and the resulting
727: eigenstates in the presence of the coupling. The eigenstates fall into
728: three classes: a) {\it Regular eigenstates} ($W>0.5$), which
729: predominantly live in the regular subspace. The remaining states,
730: which predominantly live in the chaotic subspace, are divided into two
731: classes, depending on the strength of their projection onto the regular
732: subspace compared to the equilibrium value $\Weq = \Nreg /
733: \Ntot$. This leads to b) {\it flooding eigenstates} ($0.5 \Weq < W <
734: 0.5$), and c) {\it chaotic eigenstates} ($W<0.5 \Weq$).
735: Note, that the constants 0.5 in these definitions  are arbitrary.
736: 
737: From the energy scales in the random matrix model, see
738: Fig.~\ref{fig:ladder}, we expect three qualitatively different
739: situations for the distribution of $W$:
740: 
741: i) $v\ll 1$, regular and chaotic eigenstates: In this regime the
742: regular and chaotic blocks are practically decoupled as the coupling
743: $v\Dcha$ is much smaller than the mean spacing of the chaotic basis
744: states, $v\Dcha \ll \Dcha$. Two sharp peaks are observable, one at
745: $W\approx 0$ due to the chaotic eigenstates,
746: and the other at $W \approx 1$ due to the regular eigenstates. The
747: latter peak has a smaller weight as the density of regular
748: basis states is smaller.
749: 
750: ii) $v \approx 1$, chaotic and flooding eigenstates: Here the coupling
751: $v\Dcha$ is approximately of the same order as the mean chaotic
752: spacing $\Dcha$.  All regular basis states are strongly coupled to
753: several chaotic basis states and none of the eigenstates is
754: predominantly regular.  On the other hand one has different
755: types of eigenstates as $v\Dcha < \Dreg$: Chaotic basis states, which
756: are close in energy to a regular basis state, strongly couple and thus
757: turn into flooding eigenstates. In contrast, there are many chaotic
758: basis states which are far away from any regular basis state and only
759: couple weakly. These lead to chaotic eigenstates which show
760: essentially no flooding ($W<0.5 \Weq$).
761: 
762: iii) $v \gg \Ncha/\Nreg$, flooding eigenstates: All chaotic basis
763: states are strongly coupled to the regular basis states, $v\Dcha \gg
764: \Dreg$. The resulting eigenstates equally flood the regular
765: subspace. The distribution of $W$ gets a Gaussian shape with mean
766: value $\Weq = \Nreg/\Ntot$ and a decreasing width.
767: 
768: In the transition from situation i) to ii) the two peaks of $P(W)$
769: near $W=0$ and $W=1$ broaden and move to the center. The regular peak
770: broadens faster, and at $v \approx 0.25$ its maximum disappears. At $v
771: \approx 1$ practically no eigenstates are localized in the regular
772: subspace. When moving from situation ii) to iii) the different types
773: of chaotic and flooding eigenstates transform into a single type of
774: flooding eigenstates with a similar weight $W = \Weq$ in the regular
775: subspace.
776: 
777: How do the resulting distributions depend on the ratio $\Nreg/\Ncha$?
778: First, the average of $P(W)$ is given by $\Weq = \Nreg/\Ntot
779: = 1/(1+\Ncha/\Nreg)$. Secondly, the regular peak in situation i) is independent
780: of $\Nreg/\Ncha$ apart from a trivial scaling of the normalization
781: with $\Nreg/\Ncha$. Numerically we checked that this is even true up to
782: $v\approx 1$ for the distribution with $W>0.5$ and $\Nreg/\Ncha \leq
783: 1/(8+g)$. Decreasing $\Nreg/\Ncha$ enlarges the size of the transition
784: regime between ii) and iii). In particular, the peak near $W=0$ should stay
785: there up to larger values of $v$.
786: 
787: 
788: 
789: \begin{figure*}
790:   \begin{center}
791:       \PSImagx{remapped_combined.eps}{17.2cm}
792:       \caption{(color online) 
793:                Distributions of the weights (a) $\Wnull$ and (b)
794:                 $\Weins$, taken from Figs.~\ref{fig:Weights_m0} and
795:                 \ref{fig:Weights_m1}, with $M$ rescaled to $v$
796:                 according to \eqref{veffectiveB}. (c) Result for the
797:                 random matrix (RMT) model from Fig.~\ref{fig:Weights_RMT}
798:                 on the same scale for a better comparison.
799:       \label{fig:Weights_RMT_rescaled} } 
800:   \end{center}
801: \end{figure*}
802: 
803: 
804: 
805: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
806: \section{Comparison} \label{Comparison}
807: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
808: 
809: 
810: The distribution of weights for the random matrix model, 
811: Fig.~\ref{fig:Weights_RMT}, shows a clear similarity to the
812: results obtained for the kicked system, Figs.~\ref{fig:Weights_m0} and
813: \ref{fig:Weights_m1}.  In order to obtain a quantitative comparison
814: one has to determine the relation between the coupling strength $v$ of
815: the random matrix model and the system size $M$ of the kicked
816: system. This can be deduced from Fermi's golden rule in dimensionless
817: form
818: \begin{equation}
819:   \gamma = (2 \pi)^2 \frac{\langle V^2  \rangle}{\Delta} \ , 
820:   \label{fermigoldenrule}
821: \end{equation}
822: where the decay rate $\gamma$ of a regular state to a continuum of
823: states with mean level spacing $\Delta$ is given by the variance of
824: the coupling matrix elements $V$. In the random matrix model we have
825: $\langle V^2 \rangle = (v \Dcha)^2$, $\Delta = \Dcha = 1 /\Ncha$, and
826: therefore \eqref{fermigoldenrule} implies
827: \begin{equation} \label{veffective}
828:   v = \frac{\sqrt{\gamma \Ncha}}{2 \pi} \ .  
829: \end{equation}
830: 
831: Applying this relation to the kicked system, we first note that the
832: tunnelling rate $\gamma_m$ for each torus can be determined
833: numerically \cite{Sch:PhD-BaeKetMonSch}
834: (for recent theoretical results see \cite{OniShuIkeTak2001,BroSchUll2002,PodNar2003,EltSch2005,SchEltUll2005:p,SheFisGuaReb2006}).
835: The determination of the correct value $\Ncha$ for the kicked system
836: requires a detailed discussion:
837: A regular basis state on the $m$-th torus, in the case where the tori
838: $\mfl, \mfl+1, ..., \mmax-1$ are already flooded, will couple
839: effectively to $N-\mfl M$ states for $\heff = M/N$. 
840: A change of $\mfl$ affects
841: $\Ncha$ and therefore $v$. This dependence, however, can be neglected for
842: the numerical comparison in our case: The ratio of the maximal and
843: minimal possible values of $v$ is approximately $\sqrt{ ( 1 - \heff )
844: / ( 1 - \Areg)}$. For $\heff \approx 1/10$ and $\Areg = 0.215$ this
845: gives a difference of less than 7\%. 
846: Therefore we simply use the
847: maximal value $\Ncha = N - M$ in the following.
848: 
849: For these values of $\gamma$ and $\Ncha$ in Eq.~(\ref{veffective}) 
850: the $m$-th torus of the kicked system has a coupling strength
851: \begin{equation} \label{veffectiveB}
852:   v = \frac{\sqrt{\gamma_m (1/\heff -1)}}{2 \pi} \sqrt{M} \ .  
853: \end{equation}
854: This allows for rescaling the results of the kicked system shown in
855: Figs.~\ref{fig:Weights_m0} and
856: \ref{fig:Weights_m1} from $M$ to $v$ using the values $\gamma_0 = 0.0015$
857: and $\gamma_1 = 0.030$ \cite{Sch:PhD-BaeKetMonSch}. The comparison with
858: the results from the random matrix model is shown in
859: Fig.~\ref{fig:Weights_RMT_rescaled}. The agreement is very good for
860: both tori over a wide range of coupling strengths $v$ showing the
861: universality of the flooding process. For $v > 5$, however, the
862: distribution reaches a constant width in
863: Fig.~\ref{fig:Weights_RMT_rescaled}(b), while the variance decreases
864: for the random matrix model, Fig.~\ref{fig:Weights_RMT_rescaled}(c).
865: We attribute this discrepancy to the localization of eigenstates in
866: the kicked system for $M > 1000$ \cite{HufKetOttSch2002}. 
867: As a consequence, the effective
868: number of chaotic basis states near an island saturates
869: (see the discussion after Eq.~\eqref{eq:tH}),
870: leading to an effective saturation of $v$.
871: 
872: \begin{figure}[t]
873:   \begin{center}
874:       \PSImagx{comparison_m0_144_1385.eps}{8.26cm}
875:       \vspace*{-0.25cm}
876:       \caption{(color online) 
877:                Distribution of (a) $\Wnull$ and (b) $\ln \Wnull$
878:                 for $\heff=144/1385$ (dark lines). Results of random
879:                 matrix model for $v = 0.218$ and $\Nreg/\Ncha =
880:               1/8.618$ (light lines).
881:       \label{fig:Individual_histogramsA} } 
882:   \end{center}
883: \end{figure}
884: 
885: 
886: \begin{figure}[h]
887:   \begin{center}
888:       \PSImagx{comparison_m0_1597_15360.eps}{8.26cm}
889:       \vspace*{-0.25cm}
890:       \caption{(color online) 
891:                Distribution of (a) $\Wnull$ and (b) $\ln \Wnull$
892:                 for $\heff=1597/15360$ (dark lines). Results of random
893:                 matrix model for $v = 0.726$ and $\Nreg/\Ncha =
894:               1/8.618$ (light lines).
895:       \label{fig:Individual_histogramsB} } 
896:   \end{center}
897: \end{figure}
898: 
899: 
900: 
901: In Figs.~\ref{fig:Individual_histogramsA} and
902: \ref{fig:Individual_histogramsB} we compare individual histograms 
903: for the weights $\Wnull$ for $m=0$. To visualize the low values of the
904: distributions we choose a logarithmic-linear representation in
905: Figs.~\ref{fig:Individual_histogramsA}(a) and
906: \ref{fig:Individual_histogramsB}(a). For $M=144$ one can distinguish the 
907: peak near $W=0$, due to chaotic eigenstates, from the second peak
908: caused by regular eigenstates. For $M=1597$ these two peaks have merged
909: and only a very small fraction of regular eigenstates is left. In both
910: cases the distributions agree very well with the prediction of the
911: random matrix model using $v$ according to Eq.~\eqref{veffectiveB}. To
912: resolve the peak near $W=0$ we show in
913: Figs.~\ref{fig:Individual_histogramsA}(b) and
914: \ref{fig:Individual_histogramsB}(b) the distributions of $\ln \Wnull$. 
915: Again very good agreement with the predictions of the random matrix
916: model is observed.
917: 
918: 
919: Fig.~\ref{fig:Individual_histogramsC} shows the distribution of $\ln
920: \Weins$ for $m=1$ of all eigenstates for $\heff=13/125$. We observe 
921: discrepancies at weights smaller than $10^{-3}$ in comparison to the
922: random matrix model. This difference can be explained as follows:
923: Among all the eigenstates of the kicked system there are regular
924: eigenstates localized on the torus $m=0$ which are not considered in
925: the random matrix model for $m=1$. These eigenstates have a negligible
926: overlap with the regular basis states with $m=1$ because they are
927: practically decoupled and only influence the histogram at very small
928: weights.  This is confirmed by computing the distribution, under
929: exclusion of all eigenstates with $\Wnull>0.5$.  The resulting
930: distribution matches remarkably well with the prediction of our
931: random matrix model.
932: 
933: 
934: \begin{figure}[t]
935:   \begin{center}
936:       \PSImagx{comparison_m1_13_125.eps}{8.6cm}
937:       \caption{(color online) 
938:                Distribution of $\ln \Weins$ of all eigenstates (thin
939:                 line) for $\heff=13/125$. After excluding states
940:                 with $\Wnull > 0.5$ (blue, dark line) much better
941:                 agreement with the random matrix model (red, light line) is
942:                 found.
943:       \label{fig:Individual_histogramsC} } 
944:   \end{center}
945: \end{figure}
946: 
947: 
948: 
949: 
950: 
951: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
952: \section{Fraction of regular eigenstates}\label{Fraction}
953: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
954: 
955: A more global quantity than the individual distributions $P(W)$ is the
956: fraction of regular eigenstates.  This has been studied in 
957: Ref.~\cite{BaeKetMon2005} for the total number of regular eigenstates as a
958: function of the system size. With the projection onto individual
959: regular basis states we are now able to study this fraction for each torus $m$
960: separately. For the kicked system with $M$ cells there are at most $M$
961: regular eigenstates localized on the $m$-th torus. However, during the
962: process of flooding, some of these eigenstates disappear. Thus, we
963: define the fraction $\freg$ of regular eigenstates on the $m$-th torus as
964: the number of eigenstates with weight $\Wm > 0.5$ divided by $M$. For
965: small system sizes this fraction is averaged over several different
966: phases $\chi_q$. To compare the resulting dependence on $M$ for
967: different values of $m$ and $\heff$ we determine the coupling strength
968: $v$ using Eq.~\eqref{veffectiveB}. These results are shown in
969: Fig.~\ref{fig:Fraction_regular}.
970: 
971: 
972: 
973: For the random matrix model we compute $\freg$ as the number of
974: eigenstates with $W > 0.5$ divided by the number of regular basis
975: states $\Nreg$, averaged over many realizations of the ensemble. As
976: discussed at the end of section \ref{RMModel}, the distribution $P(W)$
977: for $W>0.5$ is independent of $\Nreg/\Ncha$, apart from a trivial
978: rescaling. Therefore the resulting curve $\freg(v)$ is independent of the
979: ratio $\Nreg/\Ncha$ in contrast to the individual distributions. The
980: agreement of the fractions determined for the kicked system with the
981: random matrix curve in Fig.~\ref{fig:Fraction_regular} is very
982: good. This shows that $\freg(v)$ is a universal curve describing the
983: disappearance of regular eigenstates. For $v \leq 0.1$ the
984: fraction of regular eigenstates is larger than 98\%. 
985: For $v \geq 1$ the fraction of regular eigenstates is less than 1\% 
986: and the corresponding regular torus is completely flooded.
987: 
988: \begin{figure}[b]
989:   \begin{center}
990:       \PSImagx{fraction_remapped.eps}{8.6cm}
991:       \caption{(color online) 
992:                Fraction of regular states $\freg$ vs coupling strength $v$ 
993:                 for random
994:                 matrix model (full line) and kicked system for various
995:                 $\heff$ and $m$ (symbols),
996:                 where the system size $M$ is rescaled to $v$ according to 
997:                 Eq.~(\ref{veffectiveB}). 
998:                 Fraction of flooding eigenstates $\ffl(v)$ for the 
999:                 random matrix model (dashed line) for $\Nreg/\Ncha = 1/(8+g)$
1000:                 showing a broader transition.
1001:       \label{fig:Fraction_regular} } 
1002:   \end{center}
1003: \end{figure}
1004: 
1005: 
1006: The criterion \eqref{newcondition} for the existence of a regular
1007: eigenstate, expressed in terms of tunneling rate and Heisenberg time,
1008: can be transformed using Eqs.~\eqref{veffective} and \eqref{eq:tH}, 
1009: into the condition
1010: \begin{equation}
1011:    v < \frac{1}{2\pi} \;\;.
1012: \end{equation}
1013: The position of $v=1/(2\pi)$ 
1014: is indicated in Fig.~\ref{fig:Fraction_regular} and roughly
1015: corresponds to 93\% 
1016: of regular eigenstates still existing (by the $W >
1017: 0.5$ criterion). While in Ref.~\cite{BaeKetMon2005} condition
1018: \eqref{newcondition} for the existence of regular eigenstates
1019: was obtained from a scaling argument which does not provide a
1020: prefactor, our random matrix model analysis shows that it is 
1021: quite close to 1.
1022: 
1023: For the transition regime $1/2 \pi < v < 1$ this model shows 
1024: a decreasing probability for the existence of a regular eigenstate.
1025: For $v >1$, which implies
1026: \begin{equation}
1027:   \gamma_m > (2 \pi)^2 \frac{1}{\tH} , \label{condition_no}
1028: \end{equation}
1029: we find that almost no regular eigenstate exists on the $m$-th torus. 
1030: Thus $v=1$ defines a critical system 
1031: size $M_m$ associated with each quantized torus
1032: \begin{equation} \label{MaximalM}
1033:   M_m = \frac{ 4 \pi^2 \heff}{\gamma_m (1-\heff)} \ .
1034: \end{equation}
1035: 
1036: With the knowledge about the flooding of individual tori we can
1037: now consider the total fraction of regular eigenstates. The regular
1038: tori with larger $m$ have typically a larger tunneling rate, $\gamma_0
1039: \ll \gamma_1 \ll \ldots \ll \gamma_{\mmax-1}$. Therefore the flooding of the
1040: regular tori happens sequentially from the outside of the island as
1041: the system size increases, as found in \cite{BaeKetMon2005}. The total
1042: fraction of regular eigenstates $\cFreg$ is defined as the number of
1043: eigenstates with weights $\Wm > 0.5$ for any $m$, divided by the total
1044: number of eigenstates $N$. With Eq.~\eqref{MaximalM} we get the
1045: prediction
1046: \begin{eqnarray} \label{TotalFraction}
1047:   \cFreg (M) &=& 
1048:   \frac{M}{N}
1049:   \sum_{m=0}^{ \mmax-1} \freg \left(
1050:   \sqrt{\frac{M}{M_m}} \ \right) \ , 
1051: \end{eqnarray}
1052: where $\freg(v)$ is the universal curve from the random matrix model. 
1053: For small system sizes $M < M_m$ for all $m$ the total fraction of regular
1054: eigenstates is $\cFreg (M) = M \mmax / N \approx \Areg$, as 
1055: expected from the semiclassical eigenfunction hypothesis.
1056: Fig.~\ref{fig:Fraction_regular2} shows $\cFreg (M)$ with a
1057: succession of plateaus and drops before each critical size
1058: $M_m$. Considering that the ratio of successive $M_m$ only varies
1059: moderately, the overall behavior of $\cFreg$ is an approximately
1060: linear decrease on a logarithmic scale in $M$, explaining the
1061: observations of Ref.~\cite{BaeKetMon2005}. The agreement of
1062: Eq.~\eqref{TotalFraction} with the fraction of regular eigenstates for
1063: the kicked system for different $\heff$ as seen in
1064: Fig.~\ref{fig:Fraction_regular2} is remarkably good.
1065: 
1066: We conclude this section with the remark that due to the independence
1067: of $\freg(v)$ on the ratio $\Nreg/\Ncha$ one can obtain this universal
1068: curve by considering a simpler random matrix model, where only one
1069: regular basis state is coupled to an infinite number of chaotic basis
1070: states \cite{LeyUll96}. For this simpler model it might be possible to
1071: obtain analytical expressions for $\freg(v)$.
1072: 
1073: 
1074: \begin{figure}[t]
1075:   \begin{center}
1076:       \PSImagx{fraction2.eps}{8.6cm}
1077:       \caption{ 
1078:                Total fraction of regular states $\cFreg$ vs system size $M$
1079:                according to the prediction Eq.~\eqref{TotalFraction},
1080:                (lines) in agreement with the data for the kicked
1081:                system for $\heff \approx 1/10$ (circles) and $\heff
1082:                \approx 1/30$ (squares). 
1083:                The arrows indicate the critical system 
1084:                sizes $M_m$ according to Eq.~\eqref{MaximalM}.
1085:       \label{fig:Fraction_regular2} }
1086:   \end{center}
1087: \end{figure}
1088: 
1089: 
1090: 
1091: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1092: \section{Fraction of flooding eigenstates}\label{Fraction-flood}
1093: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1094: 
1095: The random matrix model also allows for investigating the fraction of
1096: flooding eigenstates.  While the regular eigenstates disappear with
1097: increasing coupling strength $v$, more eigenstates turn into flooding
1098: eigenstates with $0.5 \Weq < W < 0.5$.
1099: Fig.~\ref{fig:Fraction_regular} shows  the increasing fraction of these states 
1100: for the random matrix model with $\Nreg / \Ncha = 1/(8+g)$.
1101: Note, that this fraction is defined as the number
1102: of flooding eigenstates divided by the number $\Ntot$
1103: of all eigenstates.  
1104: At $v=1$ all regular
1105: eigenstates have disappeared, however, the fraction of flooding
1106: eigenstates is just 70\%. 
1107: The remaining eigenstates are chaotic, 
1108: which have no substantial weight in the regular subspace.
1109: For larger values of $v$ they turn into flooding eigenstates.
1110: This roughly happens when each chaotic basis state is coupled to at
1111: least one regular basis state, i.e.\ when $v\Dcha =\Dreg/2$, see
1112: Fig.~\ref{fig:ladder}. This gives $v=\Ncha/(2\Nreg)
1113: \approx 4.8$ which is in good agreement with the saturation observed in
1114: Fig.~\ref{fig:Fraction_regular}.
1115: This shows that the fraction of flooding eigenstates explicitly depends 
1116: on the parameter $\Nreg/\Ncha$
1117: in contrast to the fraction of regular states $\freg(v)$.
1118: 
1119: Applying this result of the random matrix model 
1120: to the kicked system 
1121: where $v=\Ncha/(2\Nreg)\approx N/(2M)$, we find
1122: using Eqs.~\eqref{veffective} and \eqref{eq:tH},
1123: that the fraction of flooding eigenstates is saturated
1124: at $\ffl=1$ for
1125: \begin{equation} \label{eq:all-flooding-states}
1126:   \gamma_m > \left( \frac{\pi}{\heff} \right)^2 \frac{1}{\tH}
1127: \end{equation}
1128: Note, that this prefactor increases in the semiclassial limit
1129: leading to a broader transition to flooding eigenstates.
1130: 
1131: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1132: \section{Summary and discussion} \label{Conclusions}
1133: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1134: 
1135: We provide a detailed quantitative description of the flooding of regular
1136: islands. By using
1137: the projection of eigenstates onto regular basis states, which defines
1138: the weights $\Wm$, the process of flooding can be described 
1139: separately for each torus. The distribution
1140: of these weights in the kicked system agrees accurately with the
1141: distribution obtained by the proposed random matrix model. 
1142: This model depends on two parameters only: the coupling
1143: strength $v$ between regular and chaotic basis states and the ratio of
1144: the number of those states $\Nreg/\Ncha$. The connection of this
1145: coupling strength with the parameters of the kicked system 
1146: is given by Eq.~\eqref{veffectiveB}.
1147: 
1148: From the random matrix model we gain the following general insights into
1149: the flooding of the $m$-th torus in terms of its tunneling rate
1150: $\gamma_m$ and the Heisenberg time $\tH$:
1151: 
1152: i) $\gamma_m < \frac{1}{\tH}$: All regular eigenstates on the $m$-th
1153: torus exist. None of the eigenstates predominantly extending over the
1154: chaotic region has substantially flooded the $m$-th torus.
1155: 
1156: ii) $\gamma_m = (2\pi)^2 \frac{1}{\tH}$: No regular eigenstates on the
1157: $m$-th torus exist. Some of the eigenstates predominantly
1158: extending over the chaotic region have substantially flooded the
1159: $m$-th torus.
1160: 
1161: iii) $\gamma_m > \left( \frac{\pi}{\heff} \right)^2 \frac{1}{\tH}$:
1162: All of the
1163: eigenstates predominantly extending over the chaotic region have
1164: substantially flooded the $m$-th torus.
1165: 
1166: What do these results imply for the applicability of
1167: the semiclassical eigenfunction hypothesis?
1168: For a fixed system size in the 
1169: semiclassical limit $\heff\rightarrow0$, 
1170: which implies a roughly exponential decrease of $\gamma_m$, one
1171: ends up in regime i), in agreement with the semiclassical
1172: eigenfunction hypothesis. In contrast,
1173: for small $\heff\neq0$ fixed and systems with $M$ cells and $M\to\infty$,
1174: one obtains a large value for $\tH\propto M$, limited by dynamical localization
1175: only.
1176: Depending on the localization length 
1177: one ends up in regime iii) for some or all tori $m$.
1178: As in our case one has $\tH \sim \heff/\gamma_0$,
1179: regime iii) is realized for all tori, i.e.\ 
1180: complete flooding of the island \cite{HufKetOttSch2002}.
1181: 
1182: The universality in the transition from i) to ii) can be seen for the
1183: fraction of regular states $\freg(v)$ localized on a given torus. 
1184: For the random matrix model this fraction does not depend on the ratio
1185: $\Nreg/\Ncha$
1186: and the agreement with the results for the kicked system is
1187: remarkably good for different quantized tori and values of
1188: $\heff$. 
1189: In contrast to the disappearance of regular eigenstates on the
1190: $m$-th torus, the transition to flooding eigenstates on this torus is
1191: much broader and extends to regime iii).
1192: 
1193: 
1194: It is also important to discuss, what these results imply
1195: for the case of a single island in a chaotic sea ($M=1$).
1196: Most commonly one is in regime i), i.e.\ 
1197: regular and chaotic eigenstates exist and only mix at accidental
1198: avoided crossings.
1199: For a sufficiently small island, compared to the size
1200: of the chaotic region, regime ii) can be reached.
1201: Here $\heff$ is small enough to quantum mechanically resolve
1202: the small regular island, but 
1203: a corresponding regular state does not exist.
1204: It is not possible, however, to get into regime iii) where
1205: all eigenstates would be flooding eigenstates: 
1206: In Eq.~\eqref{eq:all-flooding-states}
1207: we have $\heff = 1/N$ and $\tH =\Ncha\approx N$
1208: such that the right hand side is
1209: approximately $\pi^2 N$, which is always larger
1210: than the tunneling rates $\gamma_m<1$.
1211: 
1212: 
1213: 
1214: In the case of an island chain of period $p$ 
1215: embedded in a chaotic sea it might be possible to
1216: get into regime iii):
1217: In the derivation of Eq.~\eqref{eq:all-flooding-states}
1218: we now have to use $v=\Ncha/(2\Nreg)\approx N/(2p) = 1/(2p\heff)$,
1219: leading with Eqs.~\eqref{veffective} and $\Ncha \approx N = 1/\heff$
1220: to $\gamma > \pi^2/(p^2\heff)$.
1221: The right hand side can be smaller than 1 if $p$ is
1222: sufficiently large while $\heff$ is small enough
1223: to resolve the individual islands of the chain.
1224: Whether this is indeed possible in typical systems 
1225: requires further investigations.
1226: 
1227: 
1228: 
1229: This discussion shows that the semiclassical limit in generic
1230: systems with a mixed phase space, where
1231: islands of arbitrarily small size exist, is rather complicated.
1232: For example one can ask how small does $\heff$ have to be such
1233: that at least one regular state exists on a small island
1234: of size $\Areg$?
1235: Let us define the ratio $r=\heff/\Areg$
1236: The quantization condition Eq.~\eqref{eq:EBK} implies
1237: that $r < 2$ is necessary to quantum mechanically
1238: resolve the island.
1239: However, we find that the necessary ratio $r$ becomes
1240: arbitrarily small for small islands:
1241: Regime i) for $m=0$ requires $\gamma_0<1/\tH \approx \heff$. The
1242: tunneling rate $\gamma_0$ is an approximately exponentially decreasing 
1243: function $\gamma_0 \sim \exp(-C/r)$ with $C$ of the order
1244: of 1 \cite{HanOttAnt1984,FeiBaeKetRotHucBur2006}.
1245: Thus we have to fulfill 
1246: $\exp(-C/r)/r  < \Areg$,
1247: which for decreasing $\Areg$ is only possible if $r$ 
1248: is sufficiently small.
1249: 
1250: We conclude by emphasizing that the universality given by the random
1251: matrix model not only holds for the kicked system studied here,
1252: but is applicable to any system with a mixed phase space.
1253: The consequences for the semiclassical limit in the
1254: hierarchical phase--space structure of generic systems 
1255: needs much further investigation.
1256: 
1257: \vspace{0.25cm}
1258: \noindent
1259: {\bf Acknowledgements}
1260: \vspace{0.25cm}
1261: 
1262: \noindent
1263: We thank S.~Tomsovic, D.~Ullmo and
1264: H.~Weidenm\"uller for useful discussions, Lars Schilling for
1265: providing us with the data for the tunneling rates, and the Deutsche
1266: Forschungsgemeinschaft for support under contract KE 537/3-2.
1267: 
1268: 
1269: 
1270: \begin{thebibliography}{10}
1271: 
1272: \bibitem{MarMey74}
1273: L.~Markus and K.~Meyer, \emph{Generic Hamiltonian Dynamical Systems are neither
1274:   Integrable nor Chaotic}, no. 114 in Mem. Amer. Math. Soc. (American
1275:   Mathematical Society, Providence, Rhode Island, 1974).
1276: 
1277: \bibitem{Per73}
1278: I.~C. Percival, J. Phys. B \textbf{6}, L229 (1973).
1279: 
1280: \bibitem{Vor76}
1281: A.~Voros, Annales de l'Institut Henri Poincar\'e A \textbf{24}, 31 (1976).
1282: 
1283: \bibitem{Ber77b}
1284: M.~V. Berry, J. Phys. A \textbf{10}, 2083 (1977).
1285: 
1286: \bibitem{Vor79}
1287: A.~Voros, in \emph{Stochastic Behavior in Classical and Quantum Hamiltonian
1288:   Systems}, no.~93 in Lecture Notes in Physics, pages 326--333
1289:   (Springer-Verlag, Berlin, 1979).
1290: 
1291: \bibitem{Ber83}
1292: M.~V. Berry, in G.~Iooss, R.~H.~G. Hellemann and R.~Stora (eds.),
1293:   \emph{Comportement Chaotique des Syst{\`e}mes D{\'e}terministes --- Chaotic
1294:   Behaviour of Deterministic Systems}, pages 171--271 ({N}orth-{H}olland,
1295:   {A}msterdam, 1983).
1296: 
1297: \bibitem{Qerg}
1298: A.~I. Shnirelman, Usp. Math. Nauk {\bf 29} (1974) 181;
1299: S.~Zelditch, Duke. Math. J. {\bf 55} (1987) 919;
1300: Y.~{Colin de Verdi\`ere}, Commun. Math. Phys. {\bf 102} (1985) 497;
1301: B.~Helffer, A.~Martinez and D.~Robert,
1302:   Commun. Math. Phys. {\bf 109} (1987)~313;
1303: P.~G\'erard and E.~Leichtnam, Duke Math. J. {\bf 71} (1993) 559;
1304: S.~Zelditch and M.~Zworski, Commun. Math. Phys. {\bf 175} (1996) 673;
1305: A.~Bouzouina and S.~{De Bi\'evre}, Commun. Math. Phys. {\bf 178}
1306: (1996) 83;
1307: S.~{De Bi\`evre} and M.~{Degli Esposti},
1308: Ann. Inst. Henri Poincar\'e, Physique Th\'eorique {\bf 69} (1996) 1;
1309: for an introduction see A.~B\"acker, R.~Schubert and P.~Stifter,
1310: Phys. Rev. E {\bf 57} (1998) 5425.
1311: 
1312: \bibitem{BohTomUll93}
1313: O.~Bohigas, S.~Tomsovic and D.~Ullmo, Phys. Rep. \textbf{223}, 43 (1993).
1314: 
1315: \bibitem{ProRob93b}
1316: T.~Prosen and M.~Robnik, J. Phys. A \textbf{26}, 5365 (1993).
1317: 
1318: \bibitem{LiRob95b}
1319: B.~Li and M.~Robnik, J. Phys. A \textbf{28}, 4843 (1995).
1320: 
1321: \bibitem{CarVerFen98}
1322: G.~Carlo, E.~Vergini and A.~J.~Fendrik, Phys. Rev. E \textbf{57}, 5397 (1998).
1323: 
1324: \bibitem{VebRobLiu99}
1325: G.~Veble, M.~Robnik and J.~Liu, J. Phys. A \textbf{32}, 6423 (1999).
1326: 
1327: \bibitem{MarKee2005}
1328: J.~Marklof and S.~O'Keefe, Nonlinearity \textbf{18}, 277 (2005).
1329: 
1330: \bibitem{HufKetOttSch2002}
1331: L.~Hufnagel, R.~Ketzmerick, M.-F. Otto and H.~Schanz, Phys. Rev. Lett.
1332:   \textbf{89}, 154101  (2002).
1333: 
1334: \bibitem{BerBalTabVor79}
1335: M.~V. Berry, N.~L. Balazs, M.~Tabor and A.~Voros, Ann. Phys. \textbf{122}, 26
1336:   (1979).
1337: 
1338: \bibitem{DavHel81}
1339: M.~J. Davis and E.~J. Heller, J. Chem. Phys. \textbf{75}, 246 (1981).
1340: 
1341: \bibitem{ftn2}
1342: Whether these subspaces are orthogonal or only approximately orthogonal has no
1343:   consequences for the present work and for simplicity we assume orthogonality,
1344:   as e.g.\ in Ref.~\cite{BohTomUll93}.
1345: 
1346: \bibitem{BaeKetMon2005}
1347: A.~B\"acker, R.~Ketzmerick and A.~G.~Monastra, Phys. Rev. Lett. \textbf{94},
1348:   054102  (2005).
1349: 
1350: \bibitem{FeiBaeKetRotHucBur2006}
1351: J.~Feist, A.~B\"acker, R.~Ketzmerick, S.~Rotter, B.~Huckestein and
1352:   J.~Burgd\"orfer, Phys. Rev. Lett. \textbf{97}, 116804  (2006).
1353: 
1354: \bibitem{KusMosHaa1988}
1355: M.~Ku\'s, J.~Mostowski and F.~Haake, J. Phys. A \textbf{21}, L1073 (1988).
1356: 
1357: \bibitem{HaaZyc1990}
1358: F.~Haake and K.~{\.Zyczkowski}, Phys. Rev. A \textbf{42}, 1013 (1990).
1359: 
1360: \bibitem{TomUll1994}
1361: S.~Tomsovic and D.~Ullmo, Phys. Rev. E \textbf{50}, 145 (1994).
1362: 
1363: \bibitem{Bae2003}
1364: A.~B\"acker, in: {\it The Mathematical Aspects of Quantum Maps}, M. Degli
1365:   Esposti and S. Graffi (Eds.), Springer Lecture Notes in Physics \textbf{618},
1366:   91 (2003).
1367: 
1368: \bibitem{KeaMezMon2003}
1369: J.~P. Keating, F.~Mezzadri and A.~G. Monastra, J. Phys. A \textbf{36}, L53
1370:   (2003).
1371: 
1372: \bibitem{SchOttKetDit2001}
1373: H.~Schanz, M.~F. Otto, R.~Ketzmerick and T.~Dittrich, Phys. Rev. Lett.
1374:   \textbf{87}, 070601 (2001).
1375: 
1376: \bibitem{SchDitKet2005}
1377: H.~Schanz, T.~Dittrich and R.~Ketzmerick, Phys. Rev. E \textbf{71}, 026228
1378:   (2005).
1379: 
1380: \bibitem{HanBer80}
1381: J.~H. Hannay and M.~V. Berry, Physica D \textbf{1}, 267 (1980).
1382: 
1383: \bibitem{ChaShi86}
1384: S.-J. Chang and K.-J. Shi, Phys. Rev. A \textbf{34}, 7 (1986).
1385: 
1386: \bibitem{Esp93}
1387: M.~Degli~Esposti, Ann. Inst. H. Poincar\'e Phys. Th\'eor. \textbf{58}, 323
1388:   (1993).
1389: 
1390: \bibitem{DegGra03}
1391: M.~{Degli Esposti} and S.~Graffi, in: {\it The Mathematical Aspects of Quantum
1392:   Maps}, M. Degli Esposti and S. Graffi (Eds.), Springer Lecture Notes in
1393:   Physics \textbf{618}, 49 (2003).
1394: 
1395: \bibitem{Sch:PhD-BaeKetMonSch}
1396: L. Schilling, PhD thesis, TU Dresden (2006); A.~B\"acker, R.~Ketzmerick, A.~G.
1397:   Monastra and L.~Schilling, in preparation.
1398: 
1399: \bibitem{HanOttAnt1984}
1400: J.~D. Hanson, E.~Ott and T.~M. Antonsen, Phys. Rev. A \textbf{29}, 819 (1984).
1401: 
1402: \bibitem{IomFisZas2002andrefs}
1403: A. Iomin, S. Fishman, and G.~M. Zaslavsky, Phys. Rev. E {\bf 65}, 036215 (2002)
1404:   and references therein.
1405: 
1406: \bibitem{ftn}
1407: Note, that also regular states on the $m=1$ torus contribute to this peak, see
1408:   Fig.~\ref{fig:Individual_histogramsC}.
1409: 
1410: \bibitem{LeyUll96}
1411: F.~Leyvraz and D.~Ullmo, J. Phys. A \textbf{29}, 2529 (1996).
1412: 
1413: \bibitem{ZakDelBuc98}
1414: J.~Zakrzewski, D.~Delande and A.~Buchleitner, Phys. Rev. E \textbf{57}, 1458
1415:   (1998).
1416: 
1417: \bibitem{OniShuIkeTak2001}
1418: T.~Onishi, A.~Shudo, K.~S. Ikeda and K.~Takahashi, Phys. Rev. E \textbf{64},
1419:   025201  (2001).
1420: 
1421: \bibitem{BroSchUll2002}
1422: O.~Brodier, P.~Schlagheck and D.~Ullmo, Ann. Phys. \textbf{300}, 88 (2002).
1423: 
1424: \bibitem{PodNar2003}
1425: V.~A. Podolskiy and E.~E. Narimanov, Phys. Rev. Lett. \textbf{91}, 263601
1426:   (2003).
1427: 
1428: \bibitem{EltSch2005}
1429: C.~Eltschka and P.~Schlagheck, Phys. Rev. Lett. \textbf{94}, 014101 (2005).
1430: 
1431: \bibitem{SchEltUll2005:p}
1432: P.~Schlagheck, C.~Eltschka and D.~Ullmo, arXiv.org:nlin/0508024  (2005).
1433: 
1434: \bibitem{SheFisGuaReb2006}
1435: M.~Sheinman, S.~Fishman, I.~Guarneri and L.~Rebuzzini, Phys. Rev. A
1436:   \textbf{73}, 052110 (2006).
1437: 
1438: \end{thebibliography}
1439: 
1440: \end{document}
1441: 
1442: 
1443: