Turing patterns on a two-component isotropic growing system. Part 1: Homogeneous state and stability of perturbations in absence of diffusion

Abstract

The reaction-diffusion processes in a growing domain involves a dilution term that modifies the properties of the homogeneous state that, in contrast to a fixed domain, depends on time. We study how the dilution term changes the steady concentrations and modifies the stability properties of the perturbations. We propose a solution for the homogeneous state that incorporates these factors and is valid for slow variation of the size of the domain which is based on a linear approach and has been tested against numerical solutions for different types of growing: exponential, linear, quadratic and oscillatory. We prove that the deviation of the steady state is proportional to the fixed point concentration, and occurs most notably for exponential growth. Systems with linear or quadratic growth tend to recover the state that would have in absence of diffusion, whereas the oscillatory variation of the domain size produce temporal oscillations of the concentration. Regarding the Turing conditions for the apparition of spatial patterns, we study those related to stability in absence of diffusion and establish that in a growing domain it depends upon: the change in the steady state concentrations, the local change of volume that affects the concentration and the stability that the reactive system would have in absence of dilution. These conditions provide richer conditions for the emergence of patterns than those found in a fixed domain. The formal results provided in this work are verified against numerical simulations of the homogeneous state for the Brusselator and BVAM reactions and we discuss how these variations of the homogeneous state can give raise to crucial differences in the formation of Turing patterns in growing domains.