arXiv subjects

P. L. Krapivsky

Publications and source records attributed to P. L. Krapivsky.

At least 271 records · Page 15Linked to original sources

arXiv2000

Phase Transition in a Traffic Model with Passing

We investigate a traffic model in which cars either move freely with quenched intrinsic velocities or belong to clusters formed behind slower cars. In each cluster, the next-to-leading car is allowed to pass and resume free motion. The model undergoes a phase transition from a disordered phase for the high passing rate to a jammed phase for the low rate. In the disordered phase, the cluster size distribution decays exponentially in the large size limit. In the jammed phase, the cluster size distribution has a power law tail and in addition there is an infinite-size cluster. Mean-field equations, describing the model in the framework of Maxwell approximation, correctly predict the existence of phase transition and adequately describe the disordered phase; properties of the jammed phase are studied numerically.

cond-mat.stat-mech↗

arXiv1999

Fitness versus Longevity in Age-Structured Population Dynamics

We examine the dynamics of an age-structured population model in which the life expectancy of an offspring may be mutated with respect to that of the parent. While the total population of the system always reaches a steady state, the fitness and age characteristics exhibit counter-intuitive behavior as a function of the mutational bias. By analytical and numerical study of the underlying rate equations, we show that if deleterious mutations are favored, the average fitness of the population reaches a steady state, while the average population age is a decreasing function of the overall fitness. When advantageous mutations are favored, the average population fitness grows linearly with time t, while the average age is independent of fitness. For no mutational bias, the average fitness grows as t^{2/3}.

adap-org↗

arXiv1999

Particle Systems with Stochastic Passing

We study a system of particles moving on a line in the same direction. Passing is allowed and when a fast particle overtakes a slow particle, it acquires a new velocity drawn from a distribution P_0(v), while the slow particle remains unaffected. We show that the system reaches a steady state if P_0(v) vanishes at its lower cutoff; otherwise, the system evolves indefinitely.

cond-mat.stat-mech↗

arXiv1999

Scale Invariance and Lack of Self-Averaging in Fragmentation

We derive exact statistical properties of a class of recursive fragmentation processes. We show that introducing a fragmentation probability 0<p<1 leads to a purely algebraic size distribution in one dimension, P(x) ~ x^{-2p}. In d dimensions, the volume distribution diverges algebraically in the small fragment limit, P(V)\sim V^{-γ} with γ=2p^{1/d}. Hence, the entire range of exponents allowed by mass conservation is realized. We demonstrate that this fragmentation process is non-self-averaging. Specifically, the moments Y_α=\sum_i x_i^α exhibit significant fluctuations even in the thermodynamic limit.

cond-mat.stat-mech↗

arXiv1999

Multiscaling in Infinite Dimensional Collision Processes

We study relaxation properties of two-body collisions in infinite spatial dimension. We show that this process exhibits multiscaling asymptotic behavior as the underlying distribution is characterized by an infinite set of nontrivial exponents. These nonequilibrium relaxation characteristics are found to be closely related to the steady state properties of the system.

cond-mat.stat-mech↗

arXiv1999

Does Good Mutation Help You Live Longer?

We study the dynamics of an age-structured population in which the life expectancy of an offspring may be mutated with respect to that of its parent. When advantageous mutation is favored, the average fitness of the population grows linearly with time $t$, while in the opposite case the average fitness is constant. For no mutational bias, the average fitness grows as t^{2/3}. The average age of the population remains finite in all cases and paradoxically is a decreasing function of the overall population fitness.

cond-mat.stat-mech↗

arXiv1999

Capture of the Lamb: Diffusing Predators Seeking a Diffusing Prey

We study the capture of a diffusing "lamb" by diffusing "lions" in one dimension. The capture dynamics is exactly soluble by probabilistic techniques when the number of lions is very small, and is tractable by extreme statistics considerations when the number of lions is very large. However, the exact solution for the general case of three or more lions is still not known.

cond-mat.stat-mech↗

arXiv1999

Coarsening in a Driven Ising Chain with Conserved Dynamics

We study the low-temperature coarsening of an Ising chain subject to spin-exchange dynamics and a small driving force. This dynamical system reduces to a domain diffusion process, in which entire domains undergo nearest-neighbor hopping, except for the shortest domains -- dimers -- which undergo long-range hopping. This system is characterized by two independent length scales: the average domain length L(t)~t^{1/2} and the average dimer hopping distance l(t)~ t^{1/4}. As a consequence of these two scales, the density C_k(t) of domains of length k does not obey scaling. This breakdown of scaling also leads to the density of short domains decaying as t^{-5/4}, instead of the t^{-3/2} decay that would arise from pure domain diffusion.

cond-mat.stat-mech↗

arXiv1999

Comment on ``Dynamic Scaling in the Spatial Distribution of Persistent Sites''

In a recent paper (cond-mat/9901130), the spatial distribution of unvisited sites in the one-dimensional single-species annihilation process was investigated. It was claimed that this distribution is characterized by a new length scale, and that the corresponding dynamical exponent depends upon the initial concentration. We show numerically that this assertion is erroneous. Regardless of the initial concentration, this spatial distribution is characterized by the diffusive length scale.

cond-mat.stat-mech↗

arXiv1998

Maxwell Model of Traffic Flows

We investigate traffic flows using the kinetic Boltzmann equations with a Maxwell collision integral. This approach allows analytical determination of the transient behavior and the size distributions. The relaxation of the car and cluster velocity distributions towards steady state is characterized by a wide range of velocity dependent relaxation scales, $R^{1/2}<τ(v) ^{-4} Ψ(m/< m>^2)$. This distribution is primarily algebraic, $P_m\sim m^{-3/2}$, for $m\ll < m>^2$, and is exponential otherwise.

cond-mat.stat-mech↗

arXiv1998

Influence of Island Diffusion on Submonolayer Epitaxial Growth

We investigate the kinetics of submonolayer epitaxial growth which is driven by a fixed flux of monomers onto a substrate. Adatoms diffuse on the surface, leading to irreversible aggregation of islands. We also account for the effective diffusion of islands, which originates from hopping processes of their constituent adatoms, on the kinetics. When the diffusivity of an island of mass k scales as k^{-mu}, the (mean-field) Smoluchowski rate equations predicts steady behavior for 0<=mu<1, with the concentration c_k of islands of mass k varying as k^{-(3-mu)/2}. For mu>=1, a quasi-static approximation to the rate equations predicts slow continuous evolution in which the island density increases as ln t^{mu/2}. A more refined matched asymptotic expansion reveals unusual multiple-scale mass dependence for the island size distribution. Our theory also describes basic features of epitaxial growth in a more faithful model of growing circular islands. For epitaxial growth in an initial population of monomers and no external flux, a scaling approach predicts power-law island growth and a mass distribution with a behavior distinct from that of the non-zero flux system. Finally, we extend our results to one- and two-dimensional substrates. The physically-relevant latter case exhibits only logarithmic corrections compared to the mean-field predictions.

cond-mat.stat-mech↗

arXiv1998

Polydisperse Adsorption: Pattern Formation Kinetics, Fractal Properties, and Transition to Order

We investigate the process of random sequential adsorption of polydisperse particles whose size distribution exhibits a power-law dependence in the small size limit, $P(R)\sim R^{α-1}$. We reveal a relation between pattern formation kinetics and structural properties of arising patterns. We propose a mean-field theory which provides a fair description for sufficiently small $α$. When $α\to \infty$, highly ordered structures locally identical to the Apollonian packing are formed. We introduce a quantitative criterion of the regularity of the pattern formation process. When $α\gg 1$, a sharp transition from irregular to regular pattern formation regime is found to occur near the jamming coverage of standard random sequential adsorption with monodisperse size distribution.

cond-mat.stat-mech↗

arXiv1998

Slow Coarsening in an Ising Chain with Competing Interactions

We investigate the zero-temperature coarsening dynamics of a chain of Ising spins with a nearest-neighbor ferromagnetic and an nth-neighbor antiferromagnetic interactions. For sufficiently large antiferromagnetic interaction, the ground state consists of $n$ consecutive up spins followed by n down spins, etc. We show that the asymptotic coarsening into this ground state is governed by a multispecies reactive gas of elementary excitations. The basic elementary excitations are identified and each decays at a different power-law rate in time. The dominant excitations are domains of n+1 spins which diffuse freely and disappear through processes which are effectively governed by (n+1)-particle annihilation. This implies that the ground state is approached slowly with time, as t^{-1/n}.

cond-mat.stat-mech↗

arXiv1998

Domain Number Distribution in the Nonequilibrium Ising Model

We study domain distributions in the one-dimensional Ising model subject to zero-temperature Glauber and Kawasaki dynamics. The survival probability of a domain, $S(t)\sim t^{-ψ}$, and an unreacted domain, $Q_1(t)\sim t^{-δ}$, are characterized by two independent nontrivial exponents. We develop an independent interval approximation that provides close estimates for many characteristics of the domain length and number distributions including the scaling exponents.

cond-mat.stat-mech↗

arXiv1998

Fixation in a cyclic Lotka-Volterra model

We study a cyclic Lotka-Volterra model of N interacting species populating a d-dimensional lattice. In the realm of a Kirkwood approximation, a critical number of species N_c(d) above which the system fixates is determined analytically. We find N_c=5,14,23 in dimensions d=1,2,3, in remarkably good agreement with simulation results in two dimensions.

cond-mat.stat-mech↗

arXiv1997

Logarithmic Clustering in Submonolayer Epitaxial Growth

We investigate submonolayer epitaxial growth with a fixed monomer flux and irreversible aggregation of adatom islands due to their effective diffusion. When the diffusivity D_k of an island of mass k is proportional to k^{-μ}, a Smoluchowski rate equation approach predicts steady behavior for 0<μ<1, with the concentration c_k of islands of mass k varying as k^{-(3-μ)/2}. For μ>1, continuous evolution occurs in which c_k(t)~(\ln t)^{-(2k-1)μ/2}, while the total island density increases as N(t)~(\ln t)^{μ/2}. Monte Carlo simulations support these predictions.

cond-mat.stat-mech↗

arXiv1997

Mean Field Theory of Polynuclear Surface Growth

We study statistical properties of a continuum model of polynuclear surface growth on an infinite substrate. We develop a self-consistent mean-field theory which is solved to deduce the growth velocity and the extremal behavior of the coverage. Numerical simulations show that this theory gives an improved approximation for the coverage compare to the previous linear recursion relations approach. Furthermore, these two approximations provide useful upper and lower bounds for a number of characteristics including the coverage, growth velocity, and the roughness exponent.

cond-mat.stat-mech↗

arXiv1997

Alternating Kinetics of Annihilating Random Walks Near a Free Interface

The kinetics of annihilating random walks in one dimension, with the half-line x>0 initially filled, is investigated. The survival probability of the nth particle from the interface exhibits power-law decay, S_n(t)~t^{-alpha_n}, with alpha_n approximately equal to 0.225 for n=1 and all odd values of n; for all n even, a faster decay with alpha_n approximately equal to 0.865 is observed. From consideration of the eventual survival probability in a finite cluster of particles, the rigorous bound alpha_1<1/4 is derived, while a heuristic argument gives alpha_1 approximately equal to 3 sqrt{3}/8 = 0.2067.... Numerically, this latter value appears to be a stringent lower bound for alpha_1. The average position of the first particle moves to the right approximately as 1.7 t^{1/2}, with a relatively sharp and asymmetric probability distribution.

cond-mat.stat-mech↗