We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cond-mat

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Condensed Matter > Statistical Mechanics

Title: Multi-Scaling of Correlation Functions in Single Species Reaction-Diffusion Systems

Abstract: We derive the multi-fractal scaling of probability distributions of multi-particle configurations for the binary reaction-diffusion system $A+A \to \emptyset$ in $d \leq 2$ and for the ternary system $3A \to \emptyset$ in $d=1$. For the binary reaction we find that the probability $P_{t}(N, \Delta V)$ of finding $N$ particles in a fixed volume element $\Delta V$ at time $t$ decays in the limit of large time as $(\frac{\ln t}{t})^{N}(\ln t)^{-\frac{N(N-1)}{2}}$ for $d=2$ and $t^{-Nd/2}t^{-\frac{N(N-1)\epsilon}{4}+\mathcal{O}(\ep^2)}$ for $d<2$. Here $\ep=2-d$. For the ternary reaction in one dimension we find that $P_{t}(N,\Delta V) \sim (\frac{\ln t}{t})^{N/2}(\ln t)^{-\frac{N(N-1)(N-2)}{6}}$. The principal tool of our study is the dynamical renormalization group. We compare predictions of $\ep$-expansions for $P_{t}(N,\Delta V)$ for binary reaction in one dimension against exact known results. We conclude that the $\ep$-corrections of order two and higher are absent in the above answer for $P_{t}(N, \Delta V)$ for $N=1,2,3,4$. Furthermore we conjecture the absence of $\ep^2$-corrections for all values of $N$.
Comments: 10 pages, 6 figures
Subjects: Statistical Mechanics (cond-mat.stat-mech)
Journal reference: Physical Review E Vol. 73, 051103 (2006).
DOI: 10.1103/PhysRevE.73.051103
Cite as: arXiv:cond-mat/0506398 [cond-mat.stat-mech]
  (or arXiv:cond-mat/0506398v1 [cond-mat.stat-mech] for this version)

Submission history

From: Ranjiva Munasinghe [view email]
[v1] Thu, 16 Jun 2005 09:43:51 GMT (31kb)

Link back to: arXiv, form interface, contact.