Pattern formation revisited within nonequilibrium thermodynamics: Burgers'-type equation (Q2145254)

The author considers a system which is isothermal, isobaric, without external forces, and which is composed of \(n\) constituents with density \( \rho _{k}(t,x)\), velocity \(v_{k}(t,x)\), entropy \(s_{k}(t,x)\), diffusion velocity \(u_{k}\), and diffusion flux \(J_{k}\). He writes the balance equations of partial densities as \(\frac{\partial \rho _{k}}{\partial t} =-\nabla \cdot (\rho _{k}v)-\nabla \cdot J_{k}+\widehat{\rho }_{k}\), where \( \widehat{\rho }_{k}=\sum_{\alpha }v _{k\alpha }M_{k}r_{\alpha }\), \( v_{k\alpha }\) being the stoichiometric coefficients, \(M_{k}\) the molar mass, and \(r_{\alpha }\) the reaction rate of the \(\alpha \)-reaction. He cancels the velocity term in these balance equations and he adds the balance law for the entropy: \(\frac{\partial s}{\partial t}=-\nabla \cdot J_{s}+ \widehat{s}\), where \(J_{s}\) is the entropy flux and \(\widehat{s}\) is the entropy production. Focusing on a binary mixture, following his previous works and using Fick's diffusion, he ends with the equation \(\frac{\partial \rho _{1}}{\partial t}=\nabla \cdot (D_{1}\nabla \rho _{1})+K_{1}(\nabla \rho _{1})^{2}+R_{1}(\rho _{1})\), where \(D_{1}\) is Fick's coefficient, \(K_{1} \) is related to \(D_{1}\) and \(R_{1}\) is the classical reaction function. After considerations on the Hopf-Cole transformation, the author introduces the transformation \(v=-\nabla \rho \) and he derives the governing equation for \(v\) written as Burger's equation: \(\frac{\partial v}{\partial t}=D\Delta v-2Kv\cdot \nabla v+vR^{\prime }(\rho )\), \(D,K\) being positive constants and \(R\) a reaction function. He derives an implicit equation whose solution is a steady state solution to the equation satisfied by \(\rho _{1}\). He finally presents a numerical scheme for the non-stationary problem satisfied by \(\rho _{1}\), assuming that \(D_{1}\) are \(K_{1}\) are constant and considering a linear reaction function. He proves the emergence of glued parabolas and he describes their evolution with respect to time. He here refers to Turing patterns.