scientific article; zbMATH DE number 4015877
A McKean type model of bimolecular reaction \(X+Y\rightleftharpoons 2X\) is considered. Let \(N_ x(t)\) \((N_ Y(t))\) be the number of X (Y) molecules at time t. It is assumed that the total number of molecules \(N=N_ x(t)+N_ Y(t)\) does not change. Thus the contraction of this reaction involves a single random variable \(N_ x(t)\). Let \(n(t)=N_ x(t)/V\), where V is the volume, and \(P(n^ 0| nt)\) be the probability of \(n(t)=n\) given that \(n(0)=n^ 0.\) The paper gives a simple proof of Kurtz's limit theorem for the stochastic process \(n(t)\). Namely: the conditional average \(\bar n(n^ 0,t)= \sum_{n}P(n^ 0| nt)n\) satisfies in the limit \(N,V\to \infty\), with \(N/V=c\) fixed, the following differential equation \[ d\bar n/dt=k^+\bar n(c-\bar n)-k^-\bar n^ 2,\quad \bar n(0)=n^ 0. \] Moreover in this limit the scaled deviation \(q=V^{1/2}(n(t)-\bar n(n_ 0,t))\) approaches the stochastic diffusion \[ dq/dt=(d[k^+\bar n(c-\bar n)-k^-\bar n^ 2])^ q/d\bar n+(k^+n(c-\bar n)+k^+\bar n^ 2)^{1/2}\epsilon (t), \] where \(k^+,k^-\) are some constants and \(\epsilon(t)\) is a white noise of unit strength. The proof of this theorem is based on a slightly expanded version of Kac's factorization theorem, which also is obtained in this paper.
- scientific article; zbMATH DE number 4180453
- Law of large numbers and central limit theorem for linear chemical reactions with diffusion
- Density-dependent limits for a nonlinear reaction-diffusion model
- Fluctuations near homogeneous states of chemical reactions with diffusion
- Law of large numbers in the supremum norm for a chemical reaction with diffusion
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3028043)