Space-time homogenization for nonlinear diffusion

Goro Akagi, Tomoyuki Oka

Research output: Contribution to journalArticlepeer-review


The present paper is concerned with a space-time homogenization problem for nonlinear diffusion equations with periodically oscillating (in space and time) coefficients. Main results consist of a homogenization theorem (i.e., convergence of solutions as the period of oscillation goes to zero) as well as a characterization of homogenized equations. In particular, homogenized matrices are described in terms of solutions to cell-problems, which have different forms depending on the log-ratio of the spatial and temporal periods of the coefficients. At a critical ratio, the cell problem turns out to be a parabolic equation in microscopic variables (as in linear diffusion) and also involves the limit of solutions, which is a function of macroscopic variables. The latter feature stems from the nonlinearity of the equation, and moreover, some strong interplay between microscopic and macroscopic structures can explicitly be seen for the nonlinear diffusion. As for the other ratios, the cell problems are always elliptic (in micro-variable only) and do not involve any macroscopic variables, and hence, micro- and macrostructures are weakly interacting each other. Proofs of the main results are based on the two-scale convergence theory (for space-time homogenization). Furthermore, finer asymptotics of gradients, diffusion fluxes and time-derivatives with certain corrector terms are provided, and a qualitative analysis on homogenized matrices is also performed.

Original languageEnglish
Pages (from-to)386-456
Number of pages71
JournalJournal of Differential Equations
Publication statusPublished - 2023 Jun 15


  • Fast diffusion equation
  • Nonlinear diffusion
  • Periodic space-time homogenization
  • Porous media equation
  • Two-scale convergence

ASJC Scopus subject areas

  • Analysis
  • Applied Mathematics


Dive into the research topics of 'Space-time homogenization for nonlinear diffusion'. Together they form a unique fingerprint.

Cite this