1: % ****** Start of file apssamp.tex ******
2: % This file is part of the APS files in the REVTeX 4 distribution.
3: % Version 4.0 of REVTeX, August 2001
4: % Copyright (c) 2001 The American Physical Society.
5: % See the REVTeX 4 README file for restrictions and more information.
6: % TeX'ing this file requires that you have AMS-LaTeX 2.0 installed
7: % as well as the rest of the prerequisites for REVTeX 4.0
8: % See the REVTeX 4 README file
9: % It also requires running BibTeX. The commands are as follows:
10: % 1) latex apssamp.tex
11: % 2) bibtex apssamp
12: % 3) latex apssamp.tex
13: % 4) latex apssamp.tex
14: %\documentclass[preprint,showpacs,preprintnumbers,amsmath,amssymb]{revtex4}
15: % Some other (several out of many) possibilities
16: %\documentclass[preprint,aps]{revtex4}
17: %\documentclass[preprint,aps,draft]{revtex4}
18: %\documentclass[prb]{revtex4}% Physical Review B
19: % Include figure files
20: % Align table columns on decimal point
21: % bold math
22: %\nofiles
23:
24:
25: \documentclass[showpacs,preprintnumbers,12pt,onecolumn]{revtex4}
26: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27: \usepackage{amssymb}
28: \usepackage{amsfonts}
29: \usepackage{amsmath}
30: \usepackage{graphicx}
31: \usepackage{dcolumn}
32: \usepackage{bm}
33:
34: \setcounter{MaxMatrixCols}{10}
35: %TCIDATA{OutputFilter=LATEX.DLL}
36: %TCIDATA{Version=5.00.0.2557}
37: %TCIDATA{<META NAME="SaveForMode" CONTENT="1">}
38: %TCIDATA{LastRevised=Thursday, November 23, 2006 17:43:28}
39: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}
40: %TCIDATA{Language=American English}
41:
42: \input{tcilatex}
43:
44: \begin{document}
45:
46: \title{Universality in ratchets without spatial asymmetry}
47: \author{Ricardo Chac\'{o}n}
48: \affiliation{Departamento de F\'{\i}sica Aplicada, Escuela de Ingenier\'{\i}as
49: Industriales, Universidad de Extremadura, Apartado Postal 382, E-06071
50: Badajoz, Spain}
51: \date{\today}
52:
53: \begin{abstract}
54: It is demonstrated that to optimally enhance directed transport by symmetry
55: breaking of temporal forces there exists a universal force waveform which
56: allows to deduce universal scaling laws that explain previous results for a
57: great diversity of systems\ subjected to a standard biharmonic force and
58: provide a universal quantitative criterion to optimize any application of
59: the ratchet effect induced by symmetry breaking of temporal forces.
60: \end{abstract}
61:
62: \pacs{05.60.-k}
63: \maketitle
64:
65: % Force line breaks with \\
66:
67: %Lines break automatically or can be forced with \\
68:
69: % It is always \today, today,
70: % but any date may be explicitly specified
71:
72: % PACS, the Physics and Astronomy
73: % Classification Scheme.
74: %\keywords{Suggested keywords}%Use showkeys class option if keyword
75: %display desired
76:
77: \textbf{\ }Understanding the ratchet effect\ [1-4] induced by symmetry
78: breaking of temporal forces is a fundamental issue that has remained
79: unresolved for decades. While the dependence of the directed transport on
80: each of the ratchet-controlling parameters has been individually
81: investigated, there is still no general criterion to apply to the whole set
82: of these parameters to optimally control directed transport in general
83: systems without a ratchet potential [5-25]. Consider a general deterministic
84: system (classical or quantum, dissipative or non-dissipative, one- or
85: multi-dimensional) subjected to a $T$-periodic zero-mean ac force $f\left(
86: t\right) $ where a ratchet effect is induced by solely violating temporal
87: symmetries. A popular choice would be the simple case of a biharmonic force,
88: $f_{h1,h2}(t)=\epsilon _{1}har_{1}\left( \omega t+\varphi _{1}\right)
89: +\epsilon _{2}har_{2}\left( 2\omega t+\varphi _{2}\right) $, where $har_{1,2}
90: $ represents indistinctly $\sin $ or $\cos $. Clearly, the aforementioned
91: symmetries are solely the shift symmetry of the force $\left( f\left(
92: t\right) =-f\left( t+T/2\right) ,T\equiv 2\pi /\omega \right) $ and the
93: time-reversal symmetry of the system's dynamic equations. Of course, the
94: breaking of the latter symmetry implies the breaking of some time-reversal
95: symmetry of the force $\left( f\left( -t\right) =\pm f\left( t\right)
96: \right) $ in some general case, but not in all cases [19]. The analysis of
97: the breaking of these two fundamental symmetries allows to find the regions
98: of the parameter space $\left( \epsilon _{1},\epsilon _{2},\varphi
99: _{1},\varphi _{2}\right) ,\epsilon _{1}+\epsilon _{2}=const$., where the
100: ratchet effect is optimal in the sense that the average of relevant
101: observables (such as velocity and current, hereafter referred to as $%
102: \left\langle V\right\rangle $) is maximal, the remaining parameters being
103: held constant. In this Letter, it is shown that such regions are those where
104: the \textit{effective} degree of symmetry breaking is \textit{maximal}. The
105: theory arises from the observation that Curie's principle [26] implies that
106: a broken symmetry is a structurally stable situation [4]. At this point a
107: quantitative measure of the degree of symmetry breaking (DSB) is introduced,
108: on which the strength of directed transport must depend. This quantitative
109: relationship between cause (symmetry breaking) and effect (directed
110: transport) is hereafter referred to as the DSB mechanism. Also, this
111: quantitative relationship is expected to exhibit a dependence on the
112: symmetry-breaking parameters which is universal if and only if the symmetry
113: breaking takes place solely in the driving force, i.e., in the external
114: agent which is simultaneously the transport-inducing force and the
115: ratchet-inducing force. Since the ratchet effect can occur at any
116: spatio-temporal scale, such a quantitative measure of the DSB must be
117: independent of the force's amplitude and period. I define consistently the
118: DSB of the symmetries of the force $f(t)$ by the expressions%
119: \begin{eqnarray}
120: D_{s}(f) &\equiv &\left\langle \frac{-f\left( t+T/2\right) }{f(t)}%
121: \right\rangle _{T}\equiv \frac{1}{T}\int_{0}^{T}\frac{-f\left( t+T/2\right)
122: }{f(t)}dt, \notag \\
123: D_{\pm }(f) &\equiv &\left\langle \frac{\pm f\left( -t\right) }{f(t)}%
124: \right\rangle _{T}\equiv \frac{1}{T}\int_{0}^{T}\frac{\pm f\left( -t\right)
125: }{f(t)}dt, \TCItag{1}
126: \end{eqnarray}%
127: where increasing deviation of $D_{s,\pm }(f)$ from 1 (unbroken symmetry)
128: indicates an increase in the DSB. But the effectiveness of any periodic
129: zero-mean force at producing transport diminishes as the transmitted impulse
130: over a half-period is decreased while its amplitude and period are held
131: constant. In general, this means that optimal enhancement of the ratchet
132: effect is achieved when maximal \textit{effective} symmetry breaking occurs,
133: which is in turn a consequence of two reshaping-induced competing effects:
134: the increase of the degree of breaking of the force's symmetries and the
135: decrease of the transmitted impulse over a half-period, thus implying the
136: existence of a universal force waveform which optimally enhances the ratchet
137: effect. Thus, for the biharmonic force $f_{\cos ,\cos }\left( t\right)
138: =\epsilon _{1}\cos \left( \omega t+\varphi _{1}\right) +\epsilon _{2}\cos
139: \left( 2\omega t+\varphi _{2}\right) $, Eq. (1) can be put into the form $%
140: D_{s}(f_{\cos ,\cos })=1-\left( a/\pi \right) \int_{0}^{2\pi }\cos \left(
141: 2\tau +\varphi _{eff}\right) /P\left( \tau ;a,\varphi _{eff}\right) d\tau $,
142: $D_{+}(f_{\cos ,\cos })=1+\left( a/\pi \right) \int_{0}^{2\pi }\sin \left(
143: 2\tau \right) \sin \varphi _{eff}/P\left( \tau ;a,\varphi _{eff}\right)
144: d\tau $, $D_{-}(f_{\cos ,\cos })=1-\left( 1/\pi \right) \int_{0}^{2\pi }%
145: \left[ \cos \tau +a\cos \left( 2\tau \right) \cos \varphi _{eff}\right]
146: /P\left( \tau ;a,\varphi _{eff}\right) d\tau $, where $a\equiv \epsilon
147: _{2}/\epsilon _{1},\tau \equiv \omega t+\varphi _{1},\varphi \equiv \varphi
148: _{2}-2\varphi _{1},P\left( \tau ;a,\varphi _{eff}\right) \equiv \cos \tau
149: +a\cos \left( 2\tau +\varphi _{eff}\right) $. The quantity $\varphi _{eff}$
150: is hereafter referred to as the \textit{effective phase}$.$These integrals
151: diverge at the zeros of the quartic polynomial $4a^{2}x^{4}+4a\cos \varphi
152: x^{3}+\left( 1-4a^{2}\right) x^{2}-2a\cos \varphi x+a^{2}\cos ^{2}\varphi =0$%
153: , where $x\equiv \cos \tau $. After solving this algebraic equation for $x$,
154: one obtains that the three integrals diverge when $a\geqslant 1/2,\varphi
155: _{eff}=\left\{ \pi /2,3\pi /2\right\} $, and thus the DSB is maximal at
156: these parameter values for the three symmetries of the force (see Fig. 1d).
157: For these values of the effective phase, one finds that the transmitted
158: impulse over a half-period is maximal at $a=1/2$ while the biharmonic
159: force's amplitude is held constant. This means that maximal \textit{effective%
160: } symmetry breaking occurs at $a=1/2,\varphi _{eff}=\left\{ \pi /2,3\pi
161: /2\right\} $. A similar analysis of the remaining three versions of the
162: biharmonic force yields the results summarized in Table I (second column),
163: which are again the same for the three symmetries in each case. Note that
164: one could equivalently define a measure of the DSB by taking the time
165: average of the inverse quantities $-f(t)/f(t+T/2),\pm f(t)/f(-t)$: one finds
166: that the corresponding measure (1) exhibits the same qualitative behaviour
167: as a function of the symmetry-breaking parameters, and exactly the same
168: optimal values of these parameters are found to yield a maximal effective
169: DSB. This indicates that (1) provides a \textit{bona fide }measure of the
170: DSB. Remarkably, such optimal parameter values correspond to a \textit{single%
171: } optimal waveform for the four versions of the biharmonic force (see Fig.
172: 1b). The DSB mechanism implies that such a waveform is universal, i.e., it
173: corresponds to a force waveform which optimally enhances the ratchet effect
174: in any system. Consider now the case of the elliptic force $%
175: f_{ellip}(t)=\epsilon f(t;T,m,\theta )\equiv \epsilon \limfunc{sn}\left(
176: \Omega t+\Theta ;m\right) \limfunc{cn}\left( \Omega t+\Theta ;m\right) $,
177: where $\limfunc{cn}\left( \cdot ;m\right) $ and $\limfunc{sn}\left( \cdot
178: ;m\right) $ are Jacobian elliptic functions [27] of parameter $m$, $\Omega
179: \equiv 2K(m)/T,$ $\Theta \equiv K(m)\theta /\pi $, $K(m)$ is the complete
180: elliptic integral of the first kind [27], $T$ is the period of the force,
181: and $\theta $ is the (normalized) initial phase $\left( \theta \in \left[
182: 0,2\pi \right] \right) $. Fixing $\epsilon ,T$, and $\theta $, the force
183: waveform changes as the shape\textit{\ }parameter $m$ varies from 0 to 1
184: (see Fig. 1a). In this case, Eq. (1) yields $D_{s}(f_{ellip})=E(m)K^{-1}(m)%
185: \left( 1-m\right) ^{-1/2}$, where $E(m)$ is the complete elliptic integral
186: of the second kind [27] (see Fig. 1c). Physically, the motivation for this
187: choice is that $f_{ellip}(t;T,m=0,\theta )=\epsilon \sin \left( 2\pi
188: t/T+\theta \right) /2$, and that $f_{ellip}(t;T,m=1,\theta )$ vanishes,
189: i.e., in these two limits directed transport is not possible, while it is
190: expected for $0<m<1$. Thus, one may expect in general the average of any
191: relevant observable $\left\langle V\right\rangle $ to exhibit an extremum at
192: a certain critical value $m=m_{c}$ as the shape parameter $m$ is varied, the
193: remaining parameters being held constant. The DSB mechanism implies that
194: such a value $m_{c}$ is universal, i.e., it corresponds to a universal force
195: waveform which optimally enhances the ratchet effect in any system. \textit{%
196: Universality} requires that such an optimal waveform should be closely
197: related to that deduced for the case of a biharmonic force, in the sense of
198: its Fourier series. Indeed, using $f_{ellip}(t;T,m,\theta )/\epsilon
199: =\sum_{n=1}^{\infty }a_{n}(m)\sin \left[ n\left( 2\pi t/T+\theta \right) %
200: \right] $, $a_{n}(m)\equiv n\pi ^{2}m^{-1}K^{-2}(m)\func{sech}\left[ n\pi
201: K(1-m)/K(m)\right] $ one could expect the critical value $m_{c}$ to be near $%
202: m=0.983417$ since $f_{ellip}(t;T,m=0.983417,\theta )/\epsilon =a_{1}\left(
203: m=0.983417\right) \left[ \sin \left( 2\pi t/T+\theta \right) +\left(
204: 1/2\right) \sin \left( 4\pi t/T+2\theta \right) +0.178592\sin \left( 6\pi
205: t/T+3\theta \right) +...\right] $, i.e., the optimal values $\left( \epsilon
206: _{2}/\epsilon _{1}=1/2,\varphi _{2}-2\varphi _{1}=0\right) $ for the
207: biharmonic approximation of the elliptic function are recovered at $%
208: m=0.983417$ (cf. Table I, second column, and compare Figs. 1a and 1b).
209: Numerical studies of diverse systems [28] confirmed the universality and
210: accuracy of the critical value $m_{c}=0.983...$, i.e., the universality of
211: the optimal waveform. Similarly, from the Fourier series of a sawtooth-wave
212: force $f_{sawtooth}\left( t,T\right) /\epsilon =2\left[ \sin \left( 2\pi
213: t/T\right) -\left( 1/2\right) \sin \left( 4\pi t/T\right) +(1/3)\sin \left(
214: 6\pi t/T\right) -...\right] $, one recovers the optimal values $\left(
215: \epsilon _{2}/\epsilon _{1}=1/2,\varphi _{2}-2\varphi _{1}=\pi \right) $ for
216: its biharmonic approximation (cf. Table I, second column), which explains
217: the great effectiveness of this waveform in controlling directed transport
218: of magnetic flux quanta [23].
219:
220: Next, one exploits the aforementioned universality expected from the DSB
221: mechanism to deduce the dependence of $<V>$ on the symmetry-breaking
222: parameters $\left( \epsilon _{1},\epsilon _{2},\varphi _{1},\varphi
223: _{2}\right) $ of the biharmonic force $f_{h1,h2}(t)$ in leading order for
224: the usual case [5-22,24,25] of small amplitudes $\left( 1/\epsilon
225: _{1,2}\rightarrow \infty \right) $. For the sake of clarity, consider first
226: the case where the violation of the time-reversal symmetry of the system's
227: dynamic equations can be absorbed in the temporal force because dissipation
228: is negligible and the Lagrangian (Hamiltonian) of the system does not
229: contain any additional term explicitly breaking the time-reversal symmetry.
230: This means that the breaking of the time-reversal symmetry implies the
231: breaking of the force's symmetry $f\left( -t\right) =f\left( t\right) $.
232: According to the above arguments, one generally expects $\left\langle
233: V\right\rangle \sim s\left( \epsilon _{1},\epsilon _{2},\varphi _{1},\varphi
234: _{2}\right) $, where it is assumed without loss of generality that the
235: function $s$ is $k$-times piecewise continuously differentiable. From
236: MacLaurin's series, one has $s\left( \epsilon _{1},\epsilon _{2},\varphi
237: _{1},\varphi _{2}\right) =\sum_{k=0}^{\infty }\sum_{n=0}^{\infty
238: }c_{k,n}\left( \varphi _{1},\varphi _{2}\right) \epsilon _{1}^{k}\epsilon
239: _{2}^{n}$ with $c_{k,0}=c_{0,n}=0$ since the shift symmetry is never broken
240: in the case of a single harmonic function. The transformation $\epsilon
241: _{i}\rightarrow -\epsilon _{i},i=1,2,$ implies $f_{h1,h2}(t)\rightarrow
242: -f_{h1,h2}(t)$, and hence $<V>\rightarrow -<V>$. This means that $s\left(
243: \epsilon _{1},\epsilon _{2},\varphi _{1},\varphi _{2}\right) \rightarrow
244: -s\left( -\epsilon _{1},-\epsilon _{2},\varphi _{1},\varphi _{2}\right) $
245: and hence $k+n=2m+1,m=1,2,...$ . Thus, one obtains
246: \begin{equation}
247: s\left( \epsilon _{1},\epsilon _{2},\varphi _{1},\varphi _{2}\right)
248: =c_{1,2}\left( \varphi _{1},\varphi _{2}\right) \epsilon _{1}\epsilon
249: _{2}^{2}+c_{2,1}\left( \varphi _{1},\varphi _{2}\right) \epsilon
250: _{1}^{2}\epsilon _{2}+O\left( \epsilon _{1}^{2}\epsilon _{2}^{3},\epsilon
251: _{1}^{3}\epsilon _{2}^{2},\epsilon _{1}^{1}\epsilon _{2}^{4},\epsilon
252: _{1}^{4}\epsilon _{2}^{1}\right) , \tag{2}
253: \end{equation}%
254: for $\epsilon _{1,2}$ sufficiently small. Since the ratchet effect does not
255: depend on the time origin, $<V>$ must remain invariant under the
256: transformation $t\rightarrow t+t_{0},\forall t_{0}$. This transformation
257: yields $f_{h1,h2}(t)\rightarrow $ $\epsilon _{1}har_{1}\left( \omega t+%
258: \widetilde{\varphi }_{1}\right) +\epsilon _{2}har_{2}\left( 2\omega t+%
259: \widetilde{\varphi }_{2}\right) $, with the fundamental property $\widetilde{%
260: \varphi }_{2}-2\widetilde{\varphi }_{1}=\varphi _{2}-2\varphi _{1}$, i.e.,
261: the effective phase $\varphi _{2}-2\varphi _{1}$ remains invariant under
262: time translation (see Table I, third column), and hence%
263: \begin{equation}
264: \left\langle V\right\rangle \sim c_{1,2}\left( \varphi _{2}-2\varphi
265: _{1}\right) \epsilon _{1}\epsilon _{2}^{2}+c_{2,1}\left( \varphi
266: _{2}-2\varphi _{1}\right) \epsilon _{1}^{2}\epsilon _{2}. \tag{3}
267: \end{equation}%
268: The transformation $\varphi _{i}\rightarrow \varphi _{i}-\pi ,i=1,2,$
269: implies $f_{h1,h2}(t)\rightarrow -f_{h1,h2}(t)$, and hence $<V>\rightarrow
270: -<V>$ whereby $c_{1,2}\left( \varphi _{2}-2\varphi _{1}\right)
271: =-c_{1,2}\left( \varphi _{2}-2\varphi _{1}+\pi \right) ,c_{2,1}\left(
272: \varphi _{2}-2\varphi _{1}\right) =-c_{2,1}\left( \varphi _{2}-2\varphi
273: _{1}+\pi \right) $, while the transformation $\epsilon _{1}\rightarrow
274: -\epsilon _{1},\varphi _{1}\rightarrow \varphi _{1}-\pi $ maintains $%
275: \left\langle V\right\rangle $ invariant, and hence $c_{1,2}\left( \varphi
276: _{2}-2\varphi _{1}\right) =-c_{1,2}\left( \varphi _{2}-2\varphi _{1}+2\pi
277: \right) ,c_{2,1}\left( \varphi _{2}-2\varphi _{1}\right) =c_{2,1}\left(
278: \varphi _{2}-2\varphi _{1}+2\pi \right) $. The comparison of these four
279: relationships for the functions of the effective phase implies that $%
280: c_{2,1}\left( \varphi _{2}-2\varphi _{1}\right) $ is a $2\pi $-periodic
281: function while $c_{1,2}\left( \varphi _{2}-2\varphi _{1}\right) \equiv 0$.
282: Thus, Eq. (3) reduces to%
283: \begin{equation}
284: \left\langle V\right\rangle \sim \left( 1/\epsilon _{1}\right) ^{-2}\left(
285: 1/\epsilon _{2}\right) ^{-1}c_{2,1}\left( \varphi _{eff}\right) , \tag{4}
286: \end{equation}%
287: where a \textit{power law} for the dependence on the amplitudes is now
288: explicit. In this regard, it is worth noting the great similarity between
289: the present theory and the highly optimized tolerance (HOT) theory [29]
290: where a power law is generated by the actions of an external agent aiming to
291: optimize the behaviour of a system. However, we have seen above that
292: universality comes from \textit{criticality} in the present theory, while
293: for HOT systems the details matter. To obtain an explicit expression for the
294: function $c\left( \varphi _{eff}\right) $ it is useful to consider the
295: general transformation $t\rightarrow -t+t_{0},\forall t_{0}$. This
296: transformation yields $f_{h1,h2}(t)\rightarrow $ $\epsilon _{1}har_{1}\left(
297: \omega t+\widetilde{\varphi }_{1}\right) +\epsilon _{2}har_{2}\left( 2\omega
298: t+\widetilde{\varphi }_{2}\right) $ where the effective phase is no longer
299: strictly invariant but changes according to Table I (fourth column). Since
300: the transformation $t\rightarrow -t$ implies $<V>\rightarrow -<V>$ when the
301: time-reversal symmetry is unbroken, the change rules of the effective phase
302: imply that the function $c_{2,1}\left( \varphi _{eff}\right) $ has
303: necessarily a definite parity (cf. Eq. (4) and Table I, fourth column).
304: Taking into account this property and given that $c_{2,1}\left( \varphi
305: _{eff}\right) $ is assumed to be $k$-times piecewise continuously
306: differentiable, its Fourier series [30] can be approximated to leading
307: non-trivial order by a single harmonic function according to Table I (fifth
308: column). One sees that the universal scaling laws in Table I (fifth column)
309: yield $<V>=0$ when and only when both the shift symmetry and the
310: time-reversal symmetry of the system's dynamic equations (i.e., the force's
311: symmetry $f\left( -t\right) =f\left( t\right) $ in the present case) are
312: unbroken, while they yield a maximum value of $<V>$ when and only when
313: maximal \textit{effective} symmetry breaking occurs in the sense of the
314: measure (1), as predicted from the DSB mechanism. Also, that the harmonic
315: functions appearing in the universal scaling laws are independent of $har_{1}
316: $ is a consequence of the invariance of $<V>$ under time translation. As
317: expected, one finds that such universal scaling laws confirm and explain
318: previous results for a great diversity of systems [7,14,15,22,25] subjected
319: to a biharmonic force $f_{h1,h2}(t)$.
320:
321: I now discuss how the aforementioned universal scaling laws change when the
322: violation of the time-reversal symmetry of the system's dynamic equations
323: cannot be absorbed in the temporal force. This is the case when dissipation
324: [5,6,8-13,16-21,24] is not negligible. It has been demonstrated above the
325: approximate conservation of the effective phase in the sense of the change
326: rules in Table I (fourth column), and hence that $\varphi _{eff}$ is the
327: proper argument of the function $c_{2,1}\left( \varphi _{1},\varphi
328: _{2}\right) $, when the violation of the time-reversal symmetry only occurs
329: in the temporal force. Therefore, that the violation of such a symmetry is
330: also due to the presence of dissipation means that $\varphi _{eff}$ can no
331: longer be an argument of the function $c_{2,1}\left( \varphi _{1},\varphi
332: _{2}\right) $ but one has $\varphi _{eff}+\varphi _{diss}$ instead, where $%
333: \varphi _{diss}$ is hereafter referred to as the \textit{dissipation phase}.
334: Note that the additive character of the dissipation phase is a consequence
335: of the DSB mechanism. Thus, the dissipation phase quantifies the degree of
336: breaking of the time-reversal symmetry generated by dissipation. Also, the
337: DSB mechanism implies the universal properties: $\varphi _{diss}\left( \beta
338: =0\right) =0$ and that $\varphi _{diss}\left( \beta \right) $ is a
339: monotonously increasing function of $\beta $, with $\beta $ being the
340: effective dissipation parameter. For the values of $\varphi _{eff}$ yielding
341: $<V>=0$ in the absence of dissipation, i.e., those values for which the
342: temporal force does not break the time-reversal symmetry of a
343: non-dissipative system, one obtains that the maximum absolute value (i.e.,
344: 1) of the harmonic functions appearing in the universal scaling laws is
345: reached at $\varphi _{diss}=\pm \pi /2$, and hence we have the additional
346: universal property $\max_{\beta }\varphi _{diss}\left( \beta \right) =\pi /2$
347: (cf. Table I, fifth column). However, the function $\varphi _{diss}\left(
348: \beta \right) $ generally depends upon additional parameters, such as the
349: period and diverse system-dependent parameters, i.e., it is not a universal
350: function. Of course, it is generally expected that the function $%
351: <V>/har\left( \varphi _{eff}+\varphi _{diss}\right) $ should exhibit
352: monotonously decreasing behaviour as a function of $\beta $, where $har$ is
353: the corresponding harmonic function in Table I (fifth column) in each case.
354: When dissipative forces dominate inertia (the so-called overdamped regime
355: [1,4]), the breaking of the time-reversal symmetry implies the breaking of
356: the force's symmetry $f\left( -t\right) =-f\left( t\right) $ and the
357: dissipative phase reaches its limiting values $\varphi _{diss}=\pm \pi /2$.
358: Since the optimal values of the relative amplitude $\epsilon _{2}/\epsilon
359: _{1}$ and the effective phase $\varphi _{eff}$ are just the same for the
360: three symmetries of the biharmonic force, this means that the universal
361: scaling laws corresponding to the overdamped regime are those given in Table
362: I (fifth column) but with $\sin $ instead of $\cos $, and vice versa, in
363: each case. One finds that these predictions are in perfect agreement with
364: published results for a great diversity of systems [4-6,8-13,16-21,24]
365: subjected to a biharmonic force $f_{h1,h2}(t)$. Since dissipative forces and
366: randomly fluctuating forces (noise) have the same microscopic origin, it is
367: expected the effectiveness of temporal forces at generating directed
368: transport induced by the ratchet effect to be robust against moderate
369: presence of noise.
370:
371: In summary, universal scaling laws for the strength of directed transport
372: induced by symmetry breaking of temporal forces have been deduced from a
373: quantitative interpretation of Curie's principle. The present theory
374: explains in a general setting all previously published results for a great
375: diversity of systems [5-25], and provides a universal quantitative criterion
376: to optimize any application of the ratchet effect induced by symmetry
377: breaking of temporal forces.
378:
379: \begin{acknowledgments}
380: The author warmly thanks N. R. Quintero for fruitful discussions. This study
381: was supported by the Spanish MCyT and the European Regional Development Fund
382: (FEDER) program through project FIS2004-02475.
383: \end{acknowledgments}
384:
385: \bigskip
386:
387: \begin{thebibliography}{99}
388: \bibitem{1} R. P. Feynman, R. B. Leighton, and M. Sands, \textit{The Feynman
389: Lectures on Physics} (Addison-Wesley, Reading, Massachusetts, 1966),Vol. 1,
390: Ch. 46.
391:
392: \bibitem{2} P. H\"{a}nggi and R. Bartussek, in \textit{Lecture Notes in
393: Physics}, Vol. 476, p. 294, edited by J. Parisi \textit{et al}. (Springer,
394: Berlin, 1996).
395:
396: \bibitem{3} Special issue on \textit{Ratchets and Brownian Motors: Basics,
397: Experiments and Applications}, edited by H. Linke [Appl. Phys. A \textbf{75}
398: (2002)].
399:
400: \bibitem{4} P. Reimann, Phys. Rep.\textit{\ }\textbf{361}, 57 (2002).
401:
402: \bibitem{5} K. Seeger and W. Maurer, Solid State Commun. \textbf{27}, 603
403: (1978).
404:
405: \bibitem{6} A. Ajdari, D. Mukamel, L. Peliti, and J. Prost, J. Phys. I
406: France \textbf{4}, 1551 (1994).
407:
408: \bibitem{7} E. Dupont, P. B. Corkum, H. C. Liu, M. Buchanan, and Z. R.
409: Wasilewski, Phys. Rev. Lett. \textbf{74}, 3596 (1995).
410:
411: \bibitem{8} K. N. Alekseev, M. V. Erementchouk, and F. V. Kusmartsev,
412: Europhys. Lett.\textit{\ }\textbf{47}, 595 (1999).
413:
414: \bibitem{9} S. Flach, O. Yevtushenko, and Y. Zolotaryuk, Phys. Rev. Lett.%
415: \textit{\ }\textbf{84}, 2358 (2000).
416:
417: \bibitem{10} O. Yevtushenko, S. Flach, Y. Zolotaryuk, and A. A. Ovchinnikov,
418: Europhys. Lett.\textit{\ }\textbf{54}, 141 (2001).
419:
420: \bibitem{11} M. Salerno and Y. Zolotaryuk, Phys. Rev. E \textbf{65}, 56603
421: (2002).
422:
423: \bibitem{12} S. Flach, Y. Zolotaryuk, A. E. Miroshnichenko, and M. V.
424: Fistul, Phys. Rev. Lett. \textbf{88}, 184101 (2002).
425:
426: \bibitem{13} A. Engel, H. W. M\"{u}ller, P. Reimann, and A. Jung, Phys. Rev.
427: Lett.\textit{\ }\textbf{91}, 60602 (2003).
428:
429: \bibitem{14} M. Schiavoni, L. S\'{a}nchez-Palencia, F. Renzoni, and G.
430: Grynberg, Phys. Rev. Lett.\textit{\ }\textbf{90}, 94101 (2003).
431:
432: \bibitem{15} M. V. Fistul, A. E. Miroshnichenko, and S. Flach, Phys. Rev. B%
433: \textit{\ }\textbf{68}, 153107 (2003).
434:
435: \bibitem{16} J. Lehmann, S. Kohler, P. H\"{a}nggi, and A. Nitzan, J. Chem.
436: Phys.\textit{\ }\textbf{118}, 3283 (2003).
437:
438: \bibitem{17} L. Morales-Molina, N. R. Quintero, F. G. Mertens, and A. S\'{a}%
439: nchez, Phys. Rev. Lett.\textit{\ }\textbf{91}, 234102 (2003).
440:
441: \bibitem{18} A. V. Ustinov, C. Coqui, A. Kemp, Y. Zolotaryuk, and M.
442: Salerno, Phys. Rev. Lett.\textit{\ }\textbf{93}, 87001 (2004).
443:
444: \bibitem{19} R. Gommers, S. Bergamini, and F. Renzoni, Phys. Rev. Lett.%
445: \textit{\ }\textbf{95}, 73003 (2005).
446:
447: \bibitem{20} P. H\"{a}nggi, F. Marchesoni, and F. Nori, Ann. Phys. (Leipzig)%
448: \textit{\ }\textbf{14}, 51 (2005).
449:
450: \bibitem{21} X. Xie, J. Dai, and X.-C. Zhang, Phys. Rev. Lett.\textit{\ }%
451: \textbf{96}, 75005 (2006).
452:
453: \bibitem{22} I. Franco and P. Brumer, Phys. Rev. Lett.\textit{\ }\textbf{97}%
454: , 40402 (2006).
455:
456: \bibitem{23} D. Cole \textit{et al}. Nature Mat. \textbf{5}, 305 (2006).
457:
458: \bibitem{24} Y. Zolotaryuk and M. Salerno, Phys. Rev. E\textit{\ }\textbf{73}%
459: , 66621 (2006).
460:
461: \bibitem{25} S. Denisov, L. Morales-Molina, and S. Flach,
462: arXiv:cond-mat/0607558 (2006).
463:
464: \bibitem{26} P. Curie, J. de Phys. (Paris) \textbf{3}, 393 (1894).
465:
466: \bibitem{27} D. Zwillinger, \textit{Standard Mathematical Tables and Formulae%
467: } (Chapman \& Hall/CRC, London, 2003).
468:
469: \bibitem{28} R. Chac\'{o}n and N. R. Quintero, BioSystems (to be published).
470:
471: \bibitem{29} M. Newman, Nature \textbf{405}, 412 (2000); J. M. Carlson and
472: J. Doyle, Phys. Rev. Lett.\textit{\ }\textbf{84}, 2529 (2000).
473:
474: \bibitem{30} Y. Katznelson, \textit{An Introduction to Harmonic Analysis}
475: (Dover, New York, 1976).
476: \end{thebibliography}
477:
478: \bigskip
479:
480: \textbf{TABLE I}
481:
482: \begin{tabular}[t]{|c|c|c|c|c|}
483: \hline
484: $har_{1},har_{2}$ & $D_{s,\pm }$ & $t\rightarrow t+t_{0}$ & $t\rightarrow
485: -t+t_{0}$ & $\left\langle V\right\rangle $ \\ \hline
486: $\cos ,\cos $ & $\frac{\epsilon _{2}}{\epsilon _{1}}=1/2,\varphi
487: _{eff}=\left\{ \frac{\pi }{2},\frac{3\pi }{2}\right\} $ & $\widetilde{%
488: \varphi }_{eff}=\varphi _{eff}$ & $\widetilde{\varphi }_{eff}=-\varphi _{eff}
489: $ & $\sim \epsilon _{1}^{2}\epsilon _{2}\sin \varphi _{eff}$ \\ \hline
490: $\sin ,\sin $ & $\frac{\epsilon _{2}}{\epsilon _{1}}=1/2,\varphi
491: _{eff}=\left\{ 0,\pi \right\} $ & $\widetilde{\varphi }_{eff}=\varphi _{eff}$
492: & $\widetilde{\varphi }_{eff}=-\varphi _{eff}\pm \pi $ & $\sim \epsilon
493: _{1}^{2}\epsilon _{2}\cos \varphi _{eff}$ \\ \hline
494: $\sin ,\cos $ & $\frac{\epsilon _{2}}{\epsilon _{1}}=1/2,\varphi
495: _{eff}=\left\{ \frac{\pi }{2},\frac{3\pi }{2}\right\} $ & $\widetilde{%
496: \varphi }_{eff}=\varphi _{eff}$ & $\widetilde{\varphi }_{eff}=-\varphi _{eff}
497: $ & $\sim \epsilon _{1}^{2}\epsilon _{2}\sin \varphi _{eff}$ \\ \hline
498: $\cos ,\sin $ & $\frac{\epsilon _{2}}{\epsilon _{1}}=1/2,\varphi
499: _{eff}=\left\{ 0,\pi \right\} $ & $\widetilde{\varphi }_{eff}=\varphi _{eff}$
500: & $\widetilde{\varphi }_{eff}=-\varphi _{eff}\pm \pi $ & $\sim \epsilon
501: _{1}^{2}\epsilon _{2}\cos \varphi _{eff}$ \\ \hline
502: \end{tabular}
503:
504: \bigskip
505:
506: Table I. Optimal values of the relative amplitude $\epsilon _{2}/\epsilon
507: _{1}$ and the effective phase $\varphi _{eff}\equiv \varphi _{2}-2\varphi
508: _{1}$ obtained by computing the measure (1) of DSB for the three symmetries
509: of the biharmonic force (second column), change rules of the effective phase
510: under time transformations (third and fourth columns), and universal scaling
511: laws in leading order for averaged velocities and currents. Note the
512: coherence of the results in the second and fifth columns, which were
513: obtained using independent methods (recall that, without loss of generality,
514: $\epsilon _{1}+\epsilon _{2}=1$ so that $\epsilon _{1}^{2}\epsilon
515: _{2}=\left( 1-\epsilon _{2}\right) ^{2}\epsilon _{2}$, which is a function
516: having a single maximum at $\epsilon _{2}=1/3,$ and hence $\epsilon
517: _{1}=2/3,\epsilon _{2}/\epsilon _{1}=1/2.$)
518:
519: \bigskip
520:
521: \bigskip
522:
523: {\Large Figure Captions}
524:
525: \bigskip
526:
527: Figure 1. (a) Elliptic force $f_{ellip}(t)=\epsilon f(t;T,m,\theta )\equiv
528: \epsilon \limfunc{sn}\left( \Omega t+\Theta ;m\right) \limfunc{cn}\left(
529: \Omega t+\Theta ;m\right) $ vs $t/T$ and three shape parameter values, $m=0$
530: (light blue line), $m=0.983417$ (dark blue line, optimal universal
531: waveform), $m=1-10^{-6}$(light blue line), showing an increasing
532: symmetry-breaking sequence as the pulse narrows, i.e., as $m\rightarrow 1$.
533: (b) Optimal universal biharmonic force generating directed transport in one
534: direction ($f_{bihar}^{+}\left( t\right) /\epsilon =\sin (2\pi t/T)+\frac{1}{%
535: 2}\sin \left( 4\pi t/T\right) $, dark blue line) and the opposite direction (%
536: $f_{bihar}^{-}\left( t\right) /\epsilon =\sin (2\pi t/T)-(1/2)\sin \left(
537: 4\pi t/T\right) \equiv -f_{bihar}^{+}\left( t+T/2\right) /\epsilon $, red
538: line). (c) Measure of the DSB (Eq. (1)) for the elliptic force in (a), $%
539: D_{s}(f_{ellip})=E(m)K^{-1}(m)\left( 1-m\right) ^{-1/2}$ vs $m$. One sees a
540: sharp increase as $m\rightarrow 1$. (d) Measure of the DSB (Eq.(1)) for the
541: biharmonic force $f_{bihar}\left( t\right) /\epsilon =\cos \left( 2\pi
542: t/T\right) +a\cos \left( 4\pi t/T+\varphi _{eff}\right) $, $D_{s}(f_{bihar})$
543: vs $a$ for $\varphi _{eff}=\pi /2$. One sees a sharp increase as $%
544: a\rightarrow 1/2$, which is similar to that found in (c) for the elliptic
545: function.
546:
547: \end{document}
548: