Fernando A. Oliveira

ROUNDOFF-INDUCED COALESCENCE OF CHAOTIC TRAJECTORIES

Numerical experiments recently discussed in the literature show that identical nonlinear chaotic systems linked by a common noise term (or signal) may synchronize after a finite time. We study the process of synchronization as function of precision of calculations. Two generic behaviors of the average coalescence time are identified: exponential or linear. In both cases no synchronization occurs if iterations are done with {\em infinite} precision.

Normal stress distribution of rough surfaces in contact

We study numerically the stress distribution on the interface between two thick elastic media bounded by interfaces that include spatially correlated asperities. The interface roughness is described using the self-affine topography that is observed over a very wide range of scales from fractures to faults. We analyse the correlation properties of the normal stress distribution when the rough surfaces have been brought into full contact. The self affinity of the rough surfaces is described by a Hurst exponent H. We find that the normal stress field is also self affine, but with a Hurst exponent H-1. Fluctations of the normal stress are shown to be important, especially at local scales with anti-persistent correlations.

The Fluctuation-Dissipation Theorem fails for fast superdiffusion

We study anomalous diffusion for one-dimensional systems described by a generalized Langevin equation. We show that superdiffusive systems can be divided into two classes: normal and fast. For fast superdiffusion we prove that the Fluctuation-Dissipation Theorem does not hold. As a result, the system acquires an effective temperature. This effective temperature is a signature of metastability found in many complex systems such as spin-glass and granular material.

Khinchin theorem and anomalous diffusion.

A recent Letter [M. H. Lee, Phys. Rev. Lett. 98, 190601 (2007)] has called attention to the fact that irreversibility is a broader concept than ergodicity, and that therefore the Khinchin theorem [A. I. Khinchin, (Dover, New York, 1949)] may fail in some systems. In this Letter we show that for all ranges of normal and anomalous diffusion described by a generalized Langevin equation the Khinchin theorem holds.

Breaking in polymer chains. II. The Lennard‐Jones chain

Extensive simulations were performed in order to determine the conditions under which an anharmonic chain will break. The dynamics of a rectilinear chain of 100 monomers interacting via a Lennard‐Jones potential were followed by solving a set of simultaneous Langevin equations. There are two principal results from this study. First, in order for irreversible breaking to occur in a stretched chain, a bond must be extended to a length considerably greater than the length at which the restoring force is maximized. Second, the breaking rate of a bond may be expressed in terms of the product of an attempt frequency and an Arrhenius factor. While the Arrhenius factor may be satisfactorily described in terms of the height of an effective energy barrier, the attempt frequency is found to be several orders of magnitude smaller than the dominant phonon frequencies.

Non-exponential relaxation for anomalous diffusion

We study the relaxation process in normal and anomalous diffusion regimes for systems described by a generalized Langevin equation (GLE). We demonstrate the existence of a very general correlation function which describes the relaxation phenomena. Such function is even; therefore, it cannot be an exponential or a stretched exponential. However, for a proper choice of the parameters, those functions can be reproduced within certain intervals with good precision. We also show the passage from the non-Markovian to the Markovian behaviour in the normal diffusion regime. For times longer than the relaxation time, the correlation function for anomalous diffusion becomes a power law for broad-band noise.

Intermediate dynamics between Newton and Langevin.

A dynamics between Newton and Langevin formalisms is elucidated within the framework of the generalized Langevin equation. For thermal noise yielding a vanishing zero-frequency friction the corresponding non-Markovian Brownian dynamics exhibits anomalous behavior which is characterized by ballistic diffusion and accelerated transport. We also investigate the role of a possible initial correlation between the system degrees of freedom and the heat-bath degrees of freedom for the asymptotic long-time behavior of the system dynamics. As two test beds we investigate (i) the anomalous energy relaxation of free non-Markovian Brownian motion that is driven by a harmonic velocity noise and (ii) the phenomenon of a net directed acceleration in noise-induced transport of an inertial rocking Brownian motor.

Analytical results for long-time behavior in anomalous diffusion.

We investigate through a generalized Langevin formalism the phenomenon of anomalous diffusion for asymptotic times, and we generalized the concept of the diffusion exponent. A method is proposed to obtain the diffusion coefficient analytically through the introduction of a time scaling factor λ. We obtain as well an exact expression for λ for all kinds of diffusion. Moreover, we show that λ is a universal parameter determined by the diffusion exponent. The results are then compared with numerical calculations and very good agreement is observed. The method is general and may be applied to many types of stochastic problem.

Synchronization in the presence of memory

We study the effect of memory on synchronization of identical chaotic systems driven by common external noises. Our examples show that while in general the synchronization transition becomes more difficult to meet when the memory range increases, for intermediate ranges the synchronization tendency of systems can be enhanced. Generally the synchronization transition is found to depend on the memory profile and range and the ratio of noise strength to memory amplitude, which indicates a possibility of optimizing synchronization by memory. We also point out a close link between dynamics with memory and noise, and recently discovered synchronizing properties of networks with delayed interactions.

Entropy, non-ergodicity and non-Gaussian behaviour in ballistic transport

Ballistic transportation introduces new challenges in the thermodynamic properties of a gas of particles. For example, violation of mixing, ergodicity and of the fluctuation-dissipation theorem may occur, since all these processes are connected. In this work, we obtain results for all ranges of diffusion, i.e., both for subdiffusion and superdiffusion, where the bath is such that it gives origin to a colored noise. In this way we obtain the skewness and the non-Gaussian factor for the probability distribution function of the dynamical variable. We put particular emphasis on ballistic diffusion, and we demonstrate that in this case, although the second law of thermodynamics is preserved, the entropy does not reach a maximum and a non-Gaussian behavior occurs. This implies the non-applicability of the central limit theorem.