International Journal of Modern Physics C

In this work we try to verify whether the increased lifespan of trees and some lobsters, like Homarus, whose fertility increases with advancing age, can be explained by the mutation accumulation theory of biological ageing. Computer simulations of the Penna model seems to support this hypothesis, showing that it is a robust strategy of reproduction.

Cellular automata are very simple systems that can exhibit complex dynamics on its time evolution. Over the last decade there have been many applications of cellular automata to modeling of biological systems. Those applications have been stimulated by the study of complex systems which has brought many insights into the cooperative and global behavior of the biological systems. Along with this discussion we present two different applications of deterministic and also of probabilistic cellular automata that are used to model the dynamics involved in cooperative and collective behavior of the immune system. In the first example, we use a deterministic cellular automata to model the time evolution of the immune repertoire, as a network, according to the Jerne's theory. Using this model we could reproduce some recent experimental results about immunization and aging of the immune system. In the second example, we use a probabilistic cellular automata model to study the evolution of HIV infection and the onset of AIDS. The results are in excellent agreement with experimental data obtained from infected patients. Besides the examples above, other interesting applications, such as models for cancer and recurrent epidemics, are being considered in the present framework.

We have studied the aggregation of model amphiphilic molecules on a square lattice through Monte Carlo simulations via Metropolis. Amphiphilic molecules are modeled with a hydrophilic head represented by a small set of "water-loving" sites whereas the hydrophobic double-tail is represented by a second set of oil-loving sites. We have compared aggregation properties of single-stail (detergents) and double-tail (phospholipids) amphiphiles with equivalent hydrophobicity ratios. Equilibrium quantities such as average particle energy, specific heat, free amphiphile density show similar behavior for both model systems. The transition region associated with aggregation occurs at high temperatures for phospholipids as compared to detergents.

We studied different versions of the Penna bit-string model for biological ageing and found that, after many generations, the number of lineages N (maternal family names) always decays to one as a power-law N∝t^-z with an exponent z roughly equal to one. Measuring the mean correlation between the ancestor genome and those of the actual population we obtained the result that it goes to zero much earlier before the number of families goes to one, the population keeping thus its biological diversity. Considering maternal and paternal family names (doubled names) we also finished with only one pair of common ancestors. Computing the number of families of a given size as a function of the size (number of individuals the family has had during its whole existence) again a power-law decay is obtained.

We investigated two simple SOS growth models where the roughness of the profile is softened by surface relaxation or by refuse. In the first case, a particle can relax until it reaches a local minimum and this process is described by a linear equation. In the second case, a particle can be evaporated and this process is described by the KPZ equation. In this work, we have mixed both the models in the same growth process. Numerical simulations in (1+1)-dimensions show a process that for short times behaves as the surface relaxation model and for intermediate times, behaves as the refuse model. Numerical simulations indicate for this crossover a dynamical exponent z≈1.4, indicating that it is in the KPZ universality class.

We have simulated the magnetic relaxation (M(t)) and the nucleation of magnetic domains in the presence of magnetic field in thin films with anisotropy perpendicular to the film plane. We have used Monte Carlo simulations based on the two-dimensional classical Ising model including the long-range dipole–dipole and Zeeman interactions. Domains nucleated during the magnetic relaxation exhibit very rough interfaces. We analyze the roughness and the M(t) as a function of the relative strength of dipole–dipole and Zeeman terms.

Invasion percolation is studied on correlated square networks described through a site-bond model which has proven to be useful for the characterization of real heterogeneous media. It is shown how the correlation degree affects the mean front velocity, the number of islands of trapped defender fluid (which are completely surrounded by invaded elements), their size distribution and total number of steps to reach the final state. The correlation degree seems to affect the fractal dimension of the percolating cluster. A characteristic correlation length is found to exist which maximizes the mean invasion velocity.

Immiscible displacement at pore level on a three-dimensional correlated porous network is simulated allowing flow of the wetting phase along crevices of the pore walls (possibility of snap-off in throats) and advance through the centers of the pore space with different pore and throat filling conditions, leading to a cooperative filling. When these two mechanisms compete, different patterns arise. We study the effect of the correlation strength on the onset of each pattern. We do not take buoyancy forces into account.

Fiber-reinforced composites are a class of material with increasing industrial applications. Computer simulations have been used in order to understand the microscopic mechanism which can explain their mechanical behavior and several models have been introduced in the last decade. In this paper we introduce a criterion to define the brittle-ductile transition region in unidirectional fiber-reinforced composites. In order to simulate a fiber bundle, a recently introduced stochastic model is used. The results obtained with our criterion are compared with those obtained by using a self-organized criticality (SOC) approach.

We use molecular dynamics technique for simulating the zone melting process. By tuning the parameters in the molecule–molecule potential we were able to reproduce segregation and fingering at the interface.

Here we apply the Wolff single-cluster algorithm to the site-bond-correlated Ising model and study its critical dynamical behavior. We have verified that the autocorrelation time diminishes in the presence of dilution and correlation, showing that the Wolff algorithm performs even better in such situations. The critical dynamical exponents are also estimated.

Motivated by the decade-long debate over the issue of criticality supposedly observed in nuclear multifragmentation, we propose a dynamic lattice model to simulate the phenomenon. Its Ising Hamiltonian mimics a short range attractive interaction which competes with a thermal-like dissipative process. The results here presented, generated through an event-by-event analysis, are in agreement with both experiment and those produced by a percolative (nondynamic) model.

A computational study of sliding blocks on inclined surfaces is presented. Assuming that the friction coefficient μ is a function of position, the probability P(λ) for the block to slide down over a length λ is numerically calculated. Our results are consistent with recent experimental data suggesting a power-law distribution of events over a wide range of displacements when the chute angle is close to the critical one, and suggest that the variation of μ along the surface is responsible for this.

We show that for two-dimensional square Ising systems unphysical frozen states are obtained by just changing the instant of application of periodic boundary conditions during Monte Carlo simulations. The strange behavior is observed up to sample sizes currently used in literature. The anomalous results appear to be associated to the simultaneous use of type writer updating algorithms; they disappear when random access routines are implemented.

A matching algorithm for the identification of backbones in percolation problems is introduced. Using this procedure, percolation backbones are studied in two- to five-dimensional systems containing 1.7×10⁷ sites, two orders of magnitude larger than was previously possible using burning algorithms.