1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %file sodtex_a.tex
3: %\documentstyle[aps,preprint,epsf]{revtex}
4: \documentstyle[aps,pre,multicol,epsf]{revtex}
5: %\tighten
6:
7: \begin{document}
8:
9: \title{Correlations of triggering noise in driven magnetic clusters}
10:
11: \author{Bosiljka Tadi\'c$^\star $}
12:
13: \address{Jo\v{z}ef Stefan Institute,
14: P.O. Box 3000, 1001 Ljubljana, Slovenia }
15:
16: %\date\today
17:
18: \maketitle
19: \begin{abstract}
20: We show that the temporal fluctuations $\Delta H(t)$ of the
21: threshold driving field $H(t)$, which triggers an avalanche in slowly
22: driven disordered ferromagnets with many domains, exhibit long-range
23: correlations in space and time.
24: The probability distribution of the distance
25: between {\it successive} avalanches as well as the
26: distribution of trapping times of domain wall at a given point in space
27: have fractal properties with the universal scaling exponents.
28: We show how these correlations are
29: related to the scaling behavior of Barkhausen avalanches occurring by
30: magnetization reversal. We also suggest a transport equation which
31: takes into account the observed noise correlations.
32: \end{abstract}
33: \pacs{PACS numbers: 05.65.+b, 75.60.Ej, 05.40.-a }
34:
35:
36: %\narrowtext
37: \begin{multicols}{2}
38: \section{Introduction}
39: Field driven disordered ferromagnets at low temperatures exhibit
40: Barkhausen (BH) noise,
41: a very important physical phenomenon, which is used for noninvasive
42: characterization technique in commercial alloys.
43: It has been recognized that measured BH noise exhibits scale free
44: behavior when only the external field is varied for all values of
45: disorder and driving fields in real experiments
46: \cite{MC1,TOR,E4,BGD} (a short summary of the experimental data can be
47: found in Ref.\ \cite{BT-Calcutta}). This is in sharp contrast to some
48: theoretical conclusions \cite{cornell} emphasizing fine tuning of
49: the strength of disorder to a single critical value.
50: It has been also attempted to
51: understand the dynamics of domain walls which results in BH avalanches
52: in terms of the dynamics of a sandpile model \cite{GRB,Cuba} (for a recent
53: review of sandpile models see \cite{DD}) and of
54: the models of interface depinning \cite{E4,Brasil,Durin-stress,TN}.
55: It should be stressed that in BH noise the presence of disorder plays
56: an important role via pinning of domain walls. It remains to be
57: understood how the motion of domain walls is affected by the
58: pinning.
59:
60:
61: Different scaling behaviors of BH noise can be attributed
62: to the domain structure, which is related to annealing and type of
63: impurities, and nucleation and coalescence of domains, as well sa
64: varying driving conditions.
65: Annealing the samples in the applied anisotropic stress or in
66: the magnetic field leads to a characteristic structure
67: of extended system-size domains with a $180^\circ$ domain walls
68: parallel to the anisotropy axis \cite{LM,Brasil,Durin-stress}.
69: Numerical simulations with extended domain wall in two
70: dimensions \cite{TN} led to
71: the conclusion that different scaling behavior can be expected
72: in two limiting cases: (a) When disorder is weak BH response is
73: dominated by motion of a single (extended) domain wall; (b)
74: For strong disorder a multidomain structure occurs with many
75: competing domain walls. This picture is in a qualitative agreement
76: with experimental results in stress annealed
77: Fe-B-Si \cite{Brasil} and Fe-Co-B \cite{Durin-stress} alloys. Therefore
78: the two different universality classes can be related to surface and
79: bulk criticality, respectively.
80: Numerical simulations starting from an uniform ground state
81: \cite{cornell,Vives,BT}, on the other hand,
82: do not take into account extended domain walls, and thus correspond
83: to the behavior at high-disorder. In both cases, however, the origin
84: of scaling has not been fully understood.
85:
86: Recently the exact results of the random-field Ising model on the Bethe
87: lattice \cite{SD} show
88: that the avalanche distributions at a fixed driving field have
89: finite cutoffs. However, an infinite cutoff appears for a range of
90: disorder strengths $\Delta <\Delta _c$ if a
91: distribution is integrated over the hysteresis loop. The integration
92: thus involves an infinite jump in magnetization (when system size
93: $L\to \infty$) at a critical value of driving field.
94: On the other hand, for strong
95: disorder $\Delta $ above $\Delta _c$ the avalanche distributions remain
96: exponentially damped for all fields \cite{SD}. These results
97: encourage further study of the field-integrated distributions to
98: elucidate the role of driving field in the appearance of the scale-free
99: BH noise.
100:
101: In this work we study properties of the threshold driving field
102: which triggers Barkhausen
103: avalanches in the multidomain structure, which is generated in the
104: linear part of the ascending branch of hysteresis.
105: When the system is slowly driven by increasing the external magnetic
106: field $H$ with time, even by continuous field changes, there
107: exists a {\it threshold} field that equals weakest pinning force
108: of a domain wall in the system and thus starts an avalanche. In the
109: numerical experiment we can adjust
110: the driving field updates to the weakest local field in a system
111: (so-called infinitely slow driving), and thus we can examine precisely
112: the properties of the triggering noise.
113: Surprisingly, we find that time series of such field updates are
114: long-range correlated. Both Fourier spectrum of the threshold
115: field fluctuations $\Delta H(t) \equiv H(t+1)-H(t)$ and distribution of
116: distances of the successive
117: avalanches which that field triggers appear to decay with a power law.
118: The exponents are (weakly) universal in a
119: range of values of disorder where BH avalanches have a power-law
120: distributions.
121: In addition, we find that the distribution of trapping times of
122: domain wall at a given point in space exhibits scaling behavior.
123: We show how the long-range noise correlations and fractal properties
124: of the trapping time distribution are related to the observed scaling
125: behavior of BH avalanches.
126:
127: The present study is motivated by recently observed pattern
128: formation \cite{Supriya+Barma} and activity correlations along a
129: driven interface \cite{Sneppen,SOD,Mayaetal} in $1+1$ dimensions
130: stacked by random defects.
131: In contrast to these models, here we have a 2-dimensional system with
132: many interacting interfaces, and two separate time scales (compared to
133: extremal driving): slow time scale of field updates, and avalanche
134: propagation time scale between two field updates. Despite of these
135: differences, in both cases the
136: intermittent avalanche-like dynamics occurs, which we believe is
137: essential for the observed scaling behavior.
138:
139:
140:
141:
142: \section{Model and simulations}
143:
144: We consider a simple model with disorder represented by local random fields
145: $h_x$, which appears from an original disorder via coarse-graining
146: \begin{equation}
147: {\cal{H}} = -\sum_{<x,x^\prime>}J_{x,x^\prime}S_xS_{x^\prime} -
148: \sum_x (h_x+H)S_x \ ,
149: \label{Hamiltonian}
150: \end{equation}
151: where $x\equiv (x_\|,x_\bot)$ and $J_{x,x^\prime}=1$ is a constant
152: interaction between nearest-neighbor spins $S_x=\pm 1$.
153: A Gaussian distribution of $h_x$ is
154: assumed with zero mean and width $f$. (Other types of disorder have been
155: also considered \cite{BT-Calcutta,Vives,BT}). A domain wall of
156: reversed spins is created along $<11>$ direction \cite{Uli} on the
157: square lattice rotated by $\pi /4$. This model is motivated by the
158: extended domain walls in stress-annealed samples, as discussed in
159: Ref.\ \cite{TN}.
160: Periodic boundaries are applied in the direction of wall and an
161: open boundary at the opposite side of the wall. The system is driven
162: {\it globally}---updated value of the external field is applied
163: to all spins in the system. The dynamics consists of spin flips when
164: the local field exceeds zero (see below).
165:
166: As discussed in Ref.\ \cite{TN}, the fact that motion of the
167: $\langle 11 \rangle$ domain wall
168: has no energy threshold at vanishing disorder has several
169: advantages. In particular, the wall depinning occurs along
170: a line $H_c(f) \sim \kappa f$ in the $(f,H)$ plane, where $\kappa $
171: is a constant (see Fig.\ 2 in Ref.\ \cite{TN}).
172: Therefore for the domain wall along $\langle 11 \rangle $ direction
173: an {\it infinitesimally small} field $H$ is
174: sufficient to move and depin the domain wall for small disorder,
175: $H_c(f) \to 0$ for $f\to 0$.
176: This important property of
177: the model allows us to use much smaller lattice sizes compared to
178: earlier studies \cite{JR,cornell}, where a large system has to be used in
179: order to find a random field large enough to surmount the energy barrier
180: $2J$ in the case of wall along $\langle 10\rangle $ direction,
181: or the nucleation energy $4J$ in the case of uniform ground states.
182: On the other hand, by
183: increasing disorder the distance between pinning centers decreases,
184: and thus a critical disorder $f^\star$ exists at which depinning is no
185: longer possible. Instead, it becomes energetically easier to nucleate new
186: domains in the bulk. The exact value of $f^\star$
187: depends on the type of distribution \cite{TN,SD}, and is still not known
188: exactly.
189: In Ref.\ \cite{TN} it was found using a finite-size scaling analysis of
190: the averaged interface velocity and avalanche distributions
191: that for the present model $f^\star= 0.62$ within numerical error bars.
192:
193: The scaling properties of BH avalanches in low disorder regime were
194: discussed in Ref.\ \cite{TN}. Here we are interested in the high disorder
195: region, where the $\langle 11 \rangle $ interface remains pinned at
196: all fields. Nevertheless, the initial state with the $\langle 11 \rangle $
197: interface ensures that the ratio of the characteristic lengths
198: $\xi _1/L \ll 1$ holds for the applied disorder values (thus the
199: avalanche size cut-off $s_0 \ll L^2$). (The characteristic length $\xi _1$
200: represents distance between strong pinning centers \cite{JR} for given
201: strength of disorder.) On the contrary, when $\xi _1/L
202: \sim 1$, the $\langle 11 \rangle $ interface moves, meaning that the
203: selected system size $L$ is small for given strength of pinning, and thus
204: system is not in the high disorder regime.
205: Thus, having the condition $\xi _1/L \ll 1$ satisfied for each applied
206: value of disorder, we can concentrate on the effects
207: of disorder fluctuations by applying many configurations
208: of random fields at fixed $f$ and $L$ values. We use up to $L=768$
209: and up to $10^3$ disorder configurations (i.e., up to $36\times 10^6$ spins).
210:
211: %\narrowtext
212: %\begin{figure}
213: %\epsfxsize=80mm\epsffile[21 255 589 537]{bn_fig1.ps}
214: %\caption{Extended domain wall along the bottom line (shown in red), and
215: % unflipped spins (full blue color) and
216: %cluster of flipped spins developed at different stages of evolution
217: %(shown using a continuous color map [23]).
218: %Only 253 recent clusters are shown, with all older clusters shown in
219: %yellow color. $L=200$ and $f$=1.1 .
220: %}
221: %\label{fig1}
222: %\end{figure}
223: Since we are interested in the properties of the triggering noise,
224: which are related to distribution of disorder and the actual dynamics
225: of the system, we avoid any ``short-cuts'' which can speed the
226: algorithm \cite{algorithms}.
227: We apply the natural slow algorithm, which consists of the
228: following steps: Random fields are stored at each site of an $L\times L$
229: lattice; The system is searched for the minimum local field
230: $h_{min}={\rm min}(h_x^{loc})$,
231: where $h_x^{loc}\equiv \sum _{x^\prime}J_{x,x^\prime}S_{x^\prime} +h_x +H
232: \equiv h_{ir,x} +H$,
233: and then the driving field is set to exactly $h_{min}+\eta $ (we use
234: $\eta =10^{-10}$), which thus triggers an avalanche at that site;
235: A list is made consisting of the sites which may flip at the next time
236: step (neighbor sites to flipped spins); The spins on the list are
237: examined for flip and a new list is made; The process is continued until
238: no more spins can flip, then next minimum local field is
239: found.
240: It should be stressed that all spins in a single list (spin shell) are
241: updated in parallel, i.e., in a single time step, in analogy to parallel
242: update in cellular automata models. In this way we have well defined
243: time scale of avalanche evolution (internal time scale) as the number of
244: steps that the updating procedure goes before the avalanche stops.
245: On the other hand, the slow (external) time scale is set by the number
246: of driving field updates.
247:
248:
249:
250: It was shown in Ref.\ \cite{TN} that for low disorder $f<f^\star =0.62$ the
251: extended domain wall moves through the system and
252: depinns at a critical field $H_c(f)$, however, above the critical disorder
253: $f^\star $ the built-in domain wall remains pinned and many new
254: domains of reversed spins are nucleated inside the system. In this region,
255: corresponding to high disorder (or low tensile stress), a typical
256: structure of clusters which occurs in the linear part of the hysteresis is
257: shown in Fig.\ 1. The theoretical value $f^\star $ can, in principle, be
258: related to a critical value of tensile stress
259: $\sigma ^\star \sim {f{^\star}}^{\mu}$, below which the domain
260: structure changes,
261: as also observed in the experiment in Ref.\ \cite{Brasil}. More precise
262: experimental data would be necessary in order to determine the exponent $\mu$.
263: A rough estimate is that $\mu $ is given by the correlation
264: length exponent at the transition $f^\star$, $\mu \approx \nu \sim 2.3$
265: \cite{TN}.
266:
267:
268:
269:
270: \section{Fractal properties of triggering noise and domain-wall trapping}
271:
272: Time series of the magnetization jumps corresponding to the
273: individual avalanches are known as BH noise. In the high disorder region
274: either a new domain is nucleated and grown,
275: or already existing domain extended, a BH pulse is associated with
276: motion of a domain wall from one position to a new one. The area between two
277: consecutive domain wall positions corresponds to the size of BH avalanche
278: (measured as the area covered by a single BH pulse).
279: In principle, a domain growth process is not ``linear''
280: but fractal, leading to the dynamic exponent $z=1.23$, and fractal
281: dimension of avalanche $D=1.88$ \cite{TN}.
282: As usual, the dynamic exponent $z$ and the fractal dimension $D$ are
283: defined via the scaling of characteristic duration and size of avalanches
284: with the change of length scale, $\langle t\rangle _L\sim L^z$, and
285: $\langle s\rangle _L\sim L^D$, respectively.
286: Scaling properties are studied in detail (see set of scaling exponents in
287: Ref.\ \cite{TN}). In particular, in the critical region above $f^\star$
288: the distribution of size of avalanches $D(s)\sim s^{-\tau _s}$ and duration
289: $P(t) \sim t^{-\tau _t}$ scale with the exponents
290: $\tau _s=1.30$ and $\tau _t=1.47$, respectively.
291: The avalanche distributions integrated over
292: hysteresis loop for $f>f^\star$ can be scaled according to the scaling
293: form \cite{TN}
294: \begin{equation}
295: P(t,f,L)=t^{-\tau _t}{\cal{P}}(t(\delta f)^{z\nu }, tL^{-z}) \ ,
296: \label{crit_scal}
297: \end{equation}
298: where $\delta f \equiv (f/f^\star -1)$ and $(\delta f)^{-\nu }$ measures
299: the correlation length due to the critical point at $f^\star$. For instance,
300: when the condition $L\gg
301: (\delta f)^{-\nu }$ is satisfied, the avalanche cut-offs can be scaled with
302: disorder strength ({\it in the entire region of power-law behavior}),
303: leading to $f^\star =0.62$ and $\nu=2.3 $ within statistical error bars
304: (see \cite{TN} for details).
305: It should be stressed that in this region of disorder the spatial
306: extension of an avalanche is determined not only by
307: strength of pinning but also by the blocking by previous clusters
308: (see Fig.\ 1). The blocking effects dynamically alter the strength of
309: pinning and thus influence the scaling properties. This
310: illustrates a complex interplay between the disorder and dynamics,
311: in contrast to, for instance, the equilibrium clusters in the
312: random-field Ising model.
313: In order to get an insight into these complex phenomena, we study next the
314: distributions of wall trapping times and distances between the points
315: with weakest pinning, from which an avalanche is released. We find that both
316: of these distributions exhibit a scaling behavior.
317:
318: The distribution of distances between initial points of
319: the {\it successive} avalanches is
320: shown in Fig.\ 2, where we distinguish between distances measured
321: in the parallel ($x_\|$) and transverse($_\bot$) direction relative to the
322: direction of the wall. In fact, due to anisotropy of domain wall motion
323: in this case, these distributions show different scaling
324: exponents $\tau _\| =1.04 \pm 0.03$ and $\tau _\bot =0.60\pm 0.03$
325: according to
326: \begin{equation}
327: G(x_\|,x_\bot ) \sim x_\|^{-\tau _\|}{\cal{G}}(x_\bot /x_\|^{\zeta _G}) \ ,
328: \label{GG}
329: \end{equation}
330: where $\zeta _G = \tau _\| /\tau_\bot $. Notice that for small distances
331: correlations in both directions are almost equivalent. However, a crossover
332: to a distinct power-law behavior for $G(x_\bot)$ occurs at a distance
333: ($x\approx 30$ in Fig.\ 2) which is presumably related to the characteristic
334: size of avalanche at given disorder. The open boundary in the direction
335: opposite to wall leads to the cut-off at large $x_\bot$. It should be
336: stressed that spatial correlations are sensitive to distribution
337: of disorder. Therefore, here we used a large number of samples in order to
338: minimize scatter of the data due to disorder fluctuations and to clearly
339: distinguish between correlations in parallel and perpendicular directions.
340:
341:
342:
343: In the inset to Fig.\ 2 we show the Fourier spectrum of the threshold
344: field fluctuations $\Delta H(t)$ measured on the external time scale.
345: It exhibits two correlated regions
346: corresponding to earlier and later times, respectively.
347: Steeper slope in the inset to Fig.\ 2 vary from 0.6-1.03, and flatter
348: slope from 0.25 to 0.36, depending on $f$.
349: The long-range correlations exhibited in Fig.\ 1 show that the system
350: evolves in such way that its next relaxation event depends on the
351: history of the present state of the system.
352: It is interesting to note that a similar time series of the
353: fluctuation of magnetization
354: $\Delta M(t) \equiv M(t+1)-M(t)$ represents the Barkhausen noise
355: itself (looked at the external time scale). An example of the BH
356: noise signal is shown in Fig.\ 3. Note that on the external time scale
357: duration of each elementary signal is equal to one, whereas the
358: height of each elementary signal represents corresponding avalanche size.
359: Sizes of the successive avalanches are only weakly correlated in time.
360: However, a time {\it derivative} of the signal, namely $a(t) \equiv
361: d[\Delta M(t)]/dt$, representing acceleration of a domain wall
362: at each field update, shows certain correlation properties. In the inset to
363: Fig.\ 3 we show the Fourier spectrum of the numerical derivative of the
364: the signal, which is shown in main Fig.\ 3. Average slope of the curve
365: is close to one.
366:
367: It should be noted that the average number of avalanches occurring at
368: fixed strength of disorder increases with system size as $\langle n_a\rangle
369: \sim L^2/\langle s\rangle \sim L^{[2-D(2-\tau _s)]}\phi (L(\delta f)^\nu)$,
370: where $\phi (L(\delta f)^\nu)$ is unknown scaling function. This implies that
371: in the theoretical case of infinitely slow driving the average
372: size of the field jump $\langle \Delta H(t)\rangle $ decreases with $L$ as
373: $\langle \Delta H(t)\rangle \sim H_{sat}(L)/\langle n_a
374: \rangle \sim L^{-0.68}H_{sat}(L)/\phi (L(\delta f)^\nu)$, where $H_{sat}$ is
375: the saturation field. In real experimets
376: the size of field jumps are restricted by the driving frequency, however,
377: number of avalanches detected varies with the size of pick-up coil.
378: This problem requires more detailed theoretical and experimental
379: investigation \cite{Djole}.
380:
381:
382:
383: %\begin{figure}
384: %\epsfxsize=82mm\epsffile[47 68 522 581]{bn_fig2.ps}
385: %\caption{The probability that the subsequent avalanche starts at the
386: %distance $x_\|$,$x_\bot $ (bottom, top) from the point of the preceding
387: %avalanche for $L=192$ and $f=1.0$. The data are averaged over 1000 samples.
388: %Inset: Fourier spectrum of the driving field fluctuations $\Delta H(t)$
389: %taken in {\it one} ascending branch of hysteresis for $f=1.0$ and $L=192$,
390: %vs. undimensional frequency. }
391: %\label{fig2}
392: %\end{figure}
393:
394:
395:
396: Further understanding of the role of disorder in the dynamics can
397: be achieved by considering the time intervals that a domain wall resides
398: at a given point in space (trapping time $T_{trap}$).
399: We calculate the distribution of trapping times $T_{trap}$
400: of a domain wall at a given site, which is determined as the time interval
401: since the domain wall is pinned at a site $(x_\|,x_\bot )$
402: until it eventually moves away from that site. In this way, $T_{trap}$
403: measures the time interval between two successive activities at that
404: site.
405: Notice that $T_{trap}$ in this case is somewhat different from so called
406: first-return time in 1-dimensional interface \cite{Sneppen}.
407: Here we have 2-dimensional system with many interacting interfaces.
408: Another similar example is trapping of grains in rice-pile models \cite{Al}.
409: A reasonable time scale to measure $T_{trap}$ is the number of field updates
410: (external time scale), since in the zero-temperature dynamics during an
411: avalanche a spin at a given site is either fixed or flips only once.
412:
413: The trapping-time distribution
414: is shown in the inset to Fig.\ 4 for various values of disorder $f >f^\star$.
415: Between the lower cut-off $T_0$ (below which all trapping times are equally
416: probable), and an upper cut-off due to lattice size $L$,
417: the trapping time distribution shows a short region with correlations.
418: In particular, for large trapping times we can determine the slope as
419: $\tau _{trap}=2.3 \pm 0.1$.
420: One can understand that a metastable configuration of domain walls
421: corresponds to the system residing in a ``pocket'' of the fractal free
422: energy landscape. A domain wall may move from a site $(x_\|,x_\bot)$ as
423: a part of an avalanche which started in the neighborhood of that point,
424: corresponding to a local reorganization of the landscape near a shallow
425: minimum (for a given value of the external field). When the system resides
426: in a deeper minimum, however, it waits for larger driving fields
427: ({\it i.e.}, trapping time increases) or for a more global
428: reconstruction of the landscape, which occurs with smaller probability
429: compared to the one discussed above. This leads to the large slope
430: of the distribution $P(T_{trap})\sim T_{trap}^{-\tau _{trap}}$
431: at large $T_{trap}$.
432: It is interesting to note that a similar slope
433: for large trapping times was found in the case of rice-pile model \cite{Al},
434: where trapping of grains are considered, and it was argued that the slope
435: is related to the roughness exponent of the rice-pile surface
436: (see detailed analysis in Ref.\ \cite{Al}).
437: Analogies between an interface motion and rice-pile model have been
438: established in the literature \cite{Maya}. Considering the
439: avalanche exponents the analogy also applies
440: for the BH noise with an extended domain wall in the limit of low disorder
441: \cite{TN}.
442: Notice that, in contrast to
443: ricepile model, the range of power-law behavior in BH noise depends
444: not only on the system size, but also on the strength of disorder, which
445: restricts spatial extension of avalanches to $s\le s_{max}\ll L^2$ and
446: their durations to $T\le T_{max}$.
447: Therefore, in this region of disorder we have rather small range of the
448: power-law behavior. However, the scale-free
449: behavior of the distribution of trapping times can be demonstrated via
450: finite size scaling analysis when different system sizes are used. We find
451: that the following scaling form
452:
453: \begin{equation}
454: P(T_{trap},L) = L^{\alpha }{\cal{P}}(T_{trap}L^{-z_T}) \ ,
455: \label{fss}
456: \end{equation}
457: applies with the exponents which are weakly dependent on disorder
458: $\alpha =-0.4\pm 0.05$ and $z_T=1.66 \pm 0.05$ (see Fig.\ 4).
459: By increasing the system size $L$ the trapping times increase,
460: leading to larger cut-offs of the distribution. We notice that the
461: lower cut-offs $T_0$ also increase, leaving a limited correlation range.
462: However, the cut-offs scale nicely with $L$, as shown in Fig.\ 4.
463: In the inset to Fig.\ 4 we show how the trapping times distribution
464: varies with disorder.
465: By increasing disorder the cut-off $T_0$ moves towards larger values
466: and the correlation region shrinks, corresponding to a lesser
467: correlations in the dynamics, which also manifests in the decrease
468: of the characteristic avalanche size. In the limit
469: of infinitely strong disorder the dynamics becomes completely random,
470: consisting of individual spin flips which align along a local
471: random field.
472:
473:
474: It is interesting to note that unlike the avalanche distributions in
475: Eq.\ (\ref{crit_scal}), the distribution of trapping times in Eq.\
476: (\ref{fss}) as well as $G(x)$ (\ref{GG}) do not show explicit disorder
477: dependence apart from a weak dependence in the exponents. In fact, the
478: disorder effects in this case are included in the gradient of driving
479: force $\Delta F$ (i.e., the increments
480: of the external field $\Delta H$, which are adjusted to the
481: minimum local fields). For example, for the distribution of trapping times
482: we have in general
483: \begin{equation}
484: P(T_{trap},L,\Delta F) =
485: L^\alpha {\cal{P}}(T_{trap}(\Delta F)^{z_T/\lambda _F},T_{trap}L^{-z_T}) ,
486: \label{general}
487: \end{equation}
488: where the increase in the driving force $\Delta F \equiv H(t+T_{trap})-H(t)$
489: contains all the contributions due to interaction and random pinning
490: which occurred during the time interval $(t,t+T_{trap})$, i.e.,
491: $\Delta F = \sum _t^{t+T_{trap}}h_{ir,x}$, which steam from the different
492: sites in the system.
493: According to the above results in Fig.\ 2, sum of these contributions has no
494: characteristic scale.
495: Therefore we may conclude that $(\Delta F)^{-1/\lambda _F} \to \infty $.
496: Hence the first argument in the right hand side of Eq.\ (\ref{general})
497: can be neglected
498: compared to $T_{trap}/L^{z_{trap}}$, leading to the scaling form (\ref{fss}),
499: which is in agreement with the scaling plot in Fig.\ 4.
500:
501:
502: Another interesting observation regards the
503: dynamic exponent $z_T$, which scales the tail of the trapping time
504: distribution with the system size $L$. It may be related to the
505: scaling of the length of the
506: optimal path \cite{opt-path} in strongly disordered medium, $z_T=D_{OP}$.
507: In fact, the optimal path between two points can be constructed from
508: the {\it most persistent sections} of the
509: domain walls in the dynamics of BH avalanches. The length of the
510: optimal path scales with the linear distance between the end points
511: with the exponent $D_{OP}= \tau _{OP} = 1.66$
512: \cite{BT-RG,opt-path}.
513:
514:
515:
516:
517:
518: \section{Transport equation with noise correlations}
519:
520: The long-range noise correlations are shown to be relevant for the
521: scaling properties of the interface motion\cite{IF-noise}. In the
522: literature the noise correlations
523: in the interface depinning are viewed as originating from
524: another (external) dynamic processes \cite{IF-noise}.
525: In the case of Barkhausen avalanches we see that the noise correlations
526: appear as an intrinsic property of the dynamics when the system is driven
527: infinitely slowly.
528: The active section of domain wall can be represented
529: by a surface $h(x,\tau )$ which is pinned by quenched defects.
530: The transport at the site $(x,\tau )$ is described by the equation
531: \begin{equation}
532: dh/d\tau = \nu _\| \partial _\|^2 h + \nu _\bot \partial _\bot^2 h
533: + \eta (x,t,h) \ .
534: \label{transport}
535: \end{equation}
536: Here we distinguish total elapsed time $\tau $, which is defined
537: as the sum of evolution times of all individual avalanches, on one hand,
538: and $t$, which represents time scale of field updates, on the other.
539: The quenched noise $\eta (x,t,h)$ is
540: generated by varying the driving field in the steps which are adjusted
541: to the minimum local strength of pinning, as discussed above.
542: We expect that a dominate $h$-dependence of the noise $\eta $ can be
543: expressed as $\eta \sim p(x)\partial h $, where $p(x)$ represents the
544: (anisotropic) local velocity of domain wall per field rate. In addition,
545: time variation of the noise are related to the updates of the external
546: field, and thus $\eta $ varies on the external time scale only. Therefore,
547: we can write
548: \begin{equation}
549: \eta (x,t,h) \approx p(x)(a_0 + \Delta H(t))(\partial h) \; ,
550: \label{eta}
551: \end{equation}
552: where $a_0$ is a constant which
553: depends on initial configuration and the hysteresis loop properties.
554: The local interface velocity per field rate, $p(x)$, is a random variable
555: which is determined by the spatial distribution and strength of pinning
556: associated with a given value of the driving field $H$, i.e.,
557: history of the system. Therefore $p(x)$ is governed
558: by a probability distribution, which can be deduced from the properties
559: of the probability that an avalanche starts at distance $x$ from
560: the preceding avalanche, i.e.,
561: \begin{equation}
562: <p(x)p(x^\prime)> = \gamma G(x-x^\prime) \; .
563: \label{pp}
564: \end{equation}
565: It is assumed that the same type of correlations apply for the
566: successive activities during the evolution of an avalanche, while the
567: external field is constant.
568: In the previous section we found that the probability $G(x)$ exhibits
569: long-range correlations in the perpendicular and parallel direction
570: as $G(x_\|,x_\bot) \sim x_\|^{-1}x_\bot^{-0.60}$. We also notice that
571: $\Delta H(t)\sim t^{-\theta }$, where $\theta $ is finite in the case of
572: infinitely slow driving discussed above, whereas, $\theta =0$ in the
573: case of finite driving rate in small steps \cite{BT}, where
574: $\Delta H(t) =const$. Notice that
575: Eq.\ (\ref{transport}) together with Eqs.\ (\ref{eta}) and (\ref{pp})
576: leads to an effective nonlinear term of the form
577: $\gamma G(x_\|,x_\bot)(a_0+t^{-\theta })^2(\partial h)^2$, which
578: differs from the Kardar-Parisi-Zhang
579: nonlinearity \cite{KPZ} by the power-law correlations in the coefficient.
580: As discussed above, the origin of these correlations
581: lies in the the dynamically varying pinning and blocking effects
582: when a multidomain structure is slowly driven through the hysteresis loop.
583:
584:
585:
586:
587:
588:
589: The relevance of the noise correlations for the universality class of the
590: interface depinning has been discussed in
591: the literature using dynamic renormalization group (RG)
592: \cite{IF-noise}. Another example where correlations of random noise play
593: an important role is represented by the scaling properties of river
594: networks \cite{BT-RG,RN}. In that case the correlations of the type
595: $G\sim x_\|^{-2+\delta }x_\bot ^{1-\zeta}$ were assumed, where $\delta$ is
596: an expansion parameter and the anisotropy exponent $\zeta $ has to be
597: determined self-consistently at a fixed point of the RG \cite{BT-RG}.
598: In the result, the scaling exponents do depend on the range of
599: correlations $\delta$.
600: Similarly, we can expect that the long-range correlations
601: of the form given in Eqs.\ (\ref{eta}) and (\ref{pp}) can be related to the
602: universal scaling exponents of the BH noise.
603: The exponents
604: $z$, $\zeta _G$, which are defined above, and the roughness exponent
605: $\chi$, which governs behavior of the dynamic variable $h$ with the change
606: of scale, can in principle be obtained by the dynamic RG applied to
607: Eq.\ (\ref{transport}) with noise properties in Eqs.\ (\ref{eta}-\ref{pp}).
608: Then the avalanche exponents $\tau _t$ and $\tau _s$ can be deduced using
609: scaling relations in the critical region. The RG analysis of the transport
610: equation (\ref{transport}) with noise properties specified by
611: Eqs.\ (\ref{eta}-\ref{pp}) requires additional work and is left out of
612: the scope of this paper. Here we discuss only the scaling relations
613: between the exponents at RG fixed point, e.g., the dynamic exponent $z$,
614: and the avalanche exponents.
615:
616: It should be stressed that such scaling relations are not universal and that
617: they depend on the nature of the dynamic process
618: (see for instance \cite{BT-RG,RN} for the case of river networks).
619: We argue that the scaling relation derived below are valid for
620: BH avalanches in the high-disorder (or multidomain) region.
621: The rational behind these scaling relations is found in the directed
622: nature of the avalanche propagation. The evolution of an avalanche
623: (see Fig.\ 1) can be visualized as growth of a directed percolation
624: (DP) cluster projected \cite{comment} to $1+1$ dimensions, with extra
625: dimension representing time axis.
626: The fractal dimension of the equivalent DP cluster measured with respect to
627: the time axis is then $D_\| \equiv D/z$, where $D$ and $z$ are fractal
628: dimension of BH avalanche and dynamic exponent, respectively,
629: as defined above. Then $D_\|=1+\zeta _{DP} -\delta _{DP}$, where
630: $\zeta _{DP}=z/2$ is the anisotropy exponent measured with respect to
631: time axis, and $\delta _{DP}$ is the survival probability
632: exponent of the equivalent DP clusters.
633: Notice also that for the directed dynamic processes $\tau _t =D_\|$ and
634: thus $\tau _s=2-1/\tau _t$ \cite{TD}, which completes the set of
635: avalanche exponents. In addition, the roughness exponent $\chi $
636: can be related to the trapping time distribution as $\tau _{trap} =2+\chi $,
637: as noticed in Refs.\ \cite{Al,Maya}. For instance, using the numerical
638: values for $D$ and $z$ from Ref.\ \cite{TN}, the following values of
639: the avalanche
640: exponents are predicted by the above scaling relations: $\tau _t=1.52$,
641: $\tau _s=1.34$, compared with $1.47$ and $1.30$, and $\chi =0.23$,
642: obtained by direct numerical simulations.
643:
644:
645: \section{Conclusions and discussions}
646:
647: We have demonstrated numerically the existence of long-range correlations
648: in triggering noise which is intrinsic to the domain wall dynamics in
649: slowly driven disordered ferromagnets. Although the distribution of
650: disorder is initially uncorrelated, the spatio-temporal correlations
651: develop in time. The appearance of these correlations can be
652: attributed to the applied global driving in which always
653: {\it next weakest} pinning force in the system is selected, on one hand,
654: and to a finite extension of avalanches occurring between two
655: consecutive field updates, on the other.
656: Hence, attempts to reduce the domain wall dynamics to the Glauber
657: spin flips in the presence of local random fields appears to be
658: inadequate for the range of disorder where the cooperative avalanche
659: dynamics occurs.
660: We suggest an alternative transport equation which incorporates the
661: observed noise correlations.
662: The long-range correlations of triggering noise can be related to the
663: universal scaling behavior of Barkhausen avalanches and to the transport
664: properties {\it via} the fractal distribution of the domain wall
665: trapping times.
666: We believe that an analysis of the transport equation by the
667: dynamic RG will contribute to understanding of the universality
668: classes of Barkhausen avalanches by the infinitely slow driving.
669: By varying the driving conditions, however, these correlations are
670: changed. This might be the origin of different scaling behavior
671: of BH avalanches at finite driving rates, as observed in experiments
672: \cite{Durin-stress}.
673: Our results also suggest that technical procedures which
674: speed the numeric algorithms in driven disordered systems
675: should be taken with care. In particular, an algorithm which alters
676: the properties of triggering noise may lead to a different scaling
677: behavior, which is unrelated with the original problem.
678:
679:
680:
681:
682:
683:
684: \acknowledgments
685: This work was supported by the Ministry
686: of Science and Technology of the Republic of Slovenia.
687: I am grateful to Deepak Dhar for many helpful comments and suggestions.
688:
689:
690: \begin{references}
691: \bibitem[\star ]{}Electronic address: Bosiljka.Tadic@ijs.si
692:
693:
694: \bibitem{MC1} L.V. Meisel and P.J. Cote, Phys. Rev. B {\bf46}, 10822
695: (1992).
696:
697: \bibitem{TOR} G. Bertotti, G. Durin, and A. Magni, J. Appl. Phys. {\bf
698: 75}, 5490 (1994)
699:
700: \bibitem{E4}J. S. Urbach, R. C. Madison, and J. T. Markert,
701: Phys. Rev. Lett. {\bf 75}, 276 (1995).
702:
703:
704: \bibitem{BGD} Dj. Spasojevi\'c {\it et al.},
705: Phys. Rev. E {\bf 54}, 2531 (1996).
706:
707:
708: \bibitem{BT-Calcutta}B. Tadi\'c, Physica A {\bf 270}, 125 (1999).
709:
710: \bibitem{cornell}O. Perkovi\'c {\it et al.},
711: Phys. Rev. Lett. {\bf 75}, 4528 (1995); e-print cond-mat/9807336
712:
713:
714: \bibitem{GRB} O. Geoffroy and J.L. Portteseil, J.Magn. Magn. Mat. {\bf 133}, 1
715: (1994).
716:
717: \bibitem{Cuba} A. V\'asquez and O. Sotolongo-Costa, cond-mat/9903207.
718:
719: \bibitem{DD} D. Dhar, Physica A {\bf 263}, 4-25 (1999).
720:
721:
722: \bibitem{Brasil}M. Bahiana {\it et al.}, Phys. Rev. E {\bf 59},3884 (1999).
723:
724: \bibitem{Durin-stress}G. Durin and S. Zapperi, e-print cond-mat/9808224.
725:
726: \bibitem{TN}B. Tadi\'c and U. Nowak, e-print cond-mat/9903090 and
727: Phys. Rev. E (in press).
728:
729: \bibitem{LM} J. D. Livingston and W. G. Morris, J. Appl. Phys. {\bf 57},
730: 3555 (1985).
731:
732: \bibitem{Vives}E. Vives and A. Planes, Phys. Rev. E {\bf 50}, 3839 (1994).
733:
734: \bibitem{BT}B. Tadi\'c, Phys. Rev. Lett. {\bf 77}, 3843 (1996);
735: Philosophical Mag. B {\bf 77}, 277 (1998).
736:
737: \bibitem{SD}S. Sabhapandit and D. Dhar, e-print cond-mat/9905236.
738:
739: \bibitem{Supriya+Barma}S. Krishnamurthy and M. Barma, Phys. Rev. Lett.
740: {\bf 76}, 423 (1996).
741:
742: \bibitem{Sneppen}K. Sneppen and M. H. Jensen, Phys. Rev. Lett. {\bf 71},
743: 101 (1993).
744:
745: \bibitem{SOD}S. Roux and A. Hansen, J. Phys. I France {\bf 4}, 515 (1994).
746:
747: \bibitem{Mayaetal} M. Pazsuski, S. Maslov, and P. Bak, Phys. Rev. E
748: {\bf 53}, 414 (1996).
749:
750: \bibitem{Uli} A $\langle 11 \rangle $ interface was first introduced in
751: U. Nowak and K.D. Usadel, Europhys. Lett. {\bf 44}, 634 (1998).
752:
753: \bibitem{JR}H. Ji and M. O. Robbins, Phys. Rev. A {\bf 44}, 2538 (1991).
754:
755: \bibitem{color}I thank Sven L\"ubeck for providing the color map formula.
756:
757:
758: \bibitem{algorithms}M. C. Kuntz {\it et al.}, e-print cond-mat/9809122.
759:
760: \bibitem{Djole}Dj. Spasojevi\'c, private communication.
761:
762: \bibitem{Al}A. Corral, Ph. Thesis, unpublished; K. Christensen {\it et al.},
763: Phys. Rev. Lett. {\bf 77}, 107 (1996).
764:
765: \bibitem{Maya} M. Pazsuski and S. Boettcher, Phys. Rev. Lett. {\bf 77},
766: 111 (1996).
767:
768: \bibitem{BT-RG}B. Tadi\'c, Phys. Rev. E {\bf 58}, 168 (1998).
769:
770:
771: \bibitem{opt-path}M. Ceiplak {\it et al.} Phys. Rev. E (in press).
772:
773:
774: \bibitem{IF-noise} E. Medina,T. Hwa, and M. Kardar, Phys. Rev. A {\bf 39},
775: 3053 (1989).
776:
777: \bibitem{KPZ}M. Kardar, G. Parisi, and Y.C. Zhang, Phys. Rev. Lett.
778: {\bf 56}, 889 (1986); L.-H. Tang, M. Kardar, and D. Dhar, Phys. Rev.
779: Lett. {\bf 74}920 (1995).
780:
781: \bibitem{RN}J. R. Banavar {\it et al.}, Phys. Rev. Lett. {\bf 78},
782: 4522 (1997); A. Maritan {\it et al.}, Phys. Rev. E {\bf 53}, 1510 (1996).
783:
784: \bibitem{comment} Such projections are manifestly used in the problem of
785: river networks \cite{BT-RG,RN}.
786:
787: \bibitem{TD} B. Tadi\'c and D. Dhar, Phys. Rev. Lett. {\bf 79}, 1519 (1997).
788:
789: \end{references}
790:
791:
792: %\newpage
793:
794:
795: \narrowtext
796:
797: \narrowtext
798: \begin{figure}
799: \epsfxsize=80mm\epsffile[21 255 589 537]{bn_fig1.ps}
800: \caption{Extended domain wall along the bottom line (shown in red), and
801: unflipped spins (full blue color) and
802: cluster of flipped spins developed at different stages of evolution
803: (shown using a continuous color map [23]).
804: Only 253 recent clusters are shown, with all older clusters shown in
805: yellow color. $L=200$ and $f$=1.1 .
806: }
807: \label{fig1}
808: \end{figure}
809: \begin{figure}
810: \epsfxsize=82mm\epsffile[47 68 522 581]{bn_fig2.ps}
811: \caption{The probability that the subsequent avalanche starts at the
812: distance $x_\|$,$x_\bot $ (bottom, top) from the point of the preceding
813: avalanche for $L=192$ and $f=1.0$. The data are averaged over 1000 samples.
814: Inset: Fourier spectrum of the driving field fluctuations $\Delta H(t)$
815: taken in {\it one} ascending branch of hysteresis for $f=1.0$ and $L=192$,
816: vs. undimensional frequency. }
817: \label{fig2}
818: \end{figure}
819:
820:
821:
822:
823: \begin{figure}
824: \epsfxsize=82mm\epsffile[54 71 565 393]{bn_fig3.ps}
825: \caption{Sequence of Barkhausen pusles $\Delta M$ (measured in
826: number of spins) recorded on the time scale of field updates.
827: Inset: Fourier spectrum of the numerical derivative of the sequence
828: in the main figure plotted against undimensional frequency. Fitted
829: slope is $0.95\pm 0.03$.}
830: \label{fig3}
831: \end{figure}
832:
833:
834:
835: \begin{figure}
836: \epsfxsize=82mm\epsffile[47 68 522 581]{bn_fig4.ps}
837: \caption{Inset: The distribution of trapping times of a domain wall vs.
838: trapping time $T_{trap}$ measured by number of field updates.
839: The distributions for various values of disorder $f$=0.9, 1.0, 1.1, and 1.2
840: (left to right) are obtained at lattice $768\times 768$ averaged over
841: 10 samples. The data are logarithmically binned. Main: Finite size scaling
842: plot of the trapping time distribution for $f$=1.0 and the linear lattice
843: size $L$=192, 384, and 768. The data for $L=192$ are averaged over 1000
844: samples, the rest of the data over 10 samples, and are logarithmically binned.}
845: \label{fig4}
846: \end{figure}
847:
848:
849: %\end{table}
850:
851: %\end{table}
852: \end{multicols}
853:
854:
855: \end{document}
856:
857: %%%%%%%%%%%%%%%%%%%%
858:
859:
860:
861: #!/bin/csh -f
862: # this uuencoded Z-compressed .tar file created by csh script uufiles
863: # for more information, see e.g. http://xxx.lanl.gov/faq/uufaq.html
864: # if you are on a unix machine this file will unpack itself:
865: # strip off any mail header and call resulting file, e.g., bnfigs.uu
866: # (uudecode ignores these header lines and starts at begin line below)
867: # then say csh bnfigs.uu
868: # or explicitly execute the commands (generally more secure):
869: # uudecode bnfigs.uu ; uncompress bnfigs.tar.Z ;
870: # tar -xvf bnfigs.tar
871: # on some non-unix (e.g. VAX/VMS), first use an editor to change the
872: # filename in "begin" line below to bnfigs.tar_Z , then execute
873: # uudecode bnfigs.uu
874: # compress -d bnfigs.tar_Z
875: # tar -xvf bnfigs.tar
876: #
877: uudecode $0
878: chmod 644 bnfigs.tar.Z
879: zcat bnfigs.tar.Z | tar -xvf -
880: rm $0 bnfigs.tar.Z
881: exit
882:
883: