arXiv Analytics

Sign in

arXiv:2402.16451 [math.AP]AbstractReferencesReviewsResources

The Neumann sieve problem revisited

Andrii Khrabustovskyi

Published 2024-02-26, updated 2024-04-04Version 2

Let $\Omega\subset\mathbb{R}^n$ be a domain, $\Gamma$ be a hyperplane intersecting it. Let $\varepsilon>0$, and $\Omega_\varepsilon=\Omega\setminus\overline{\Sigma_\varepsilon}$, where $\Sigma_\varepsilon$ ("sieve") is an $\varepsilon$-neighbourhood of $\Gamma$ punctured by many narrow passages. When $\varepsilon\to0$, the number of passages tends to infinity, while the diameters of their cross-sections tend to zero. For the case of identical straight periodically distributed and appropriately scaled passages T. Del Vecchio (1987) proved that the Neumann Laplacian on $\Omega_\varepsilon$ converges in a strong resolvent sense to the Laplacian on $\Omega\setminus\Gamma$ subject to the so-called $\delta'$-conditions on $\Gamma$. We will refine this result by deriving estimates on the rate of convergence in terms of various operator norms, and providing the estimate for the distance between the spectra. The assumptions we impose on the passages are rather general. For $n=2$ the results of T. Del Vecchio are not complete, some cases remain as open problems, and in this work we will fill these gaps.

Related articles: Most relevant | Search more
arXiv:2209.08775 [math.AP] (Published 2022-09-19)
Operator estimates for Neumann sieve problem
arXiv:2106.10216 [math.AP] (Published 2021-06-18)
Operator estimates for homogenization of the Robin Laplacian in a perforated domain
arXiv:math/0605769 [math.AP] (Published 2006-05-30)
The Neumann sieve problem and dimensional reduction: a multiscale approach