Skip to main content
10 events
when toggle format what by license comment
Aug 14, 2023 at 21:06 history edited Connor Mooney CC BY-SA 4.0
added 1 character in body
Aug 14, 2023 at 21:05 comment added Connor Mooney @Jochen: True. To be precise I've added that we consider Dirichlet boundary conditions, $L = \Delta$, and smooth domains. As you point out, the argument works in more general scenarios provided the eigenfunctions are in $C^2(\Omega) \cap C^1(\overline{\Omega})$.
Aug 14, 2023 at 21:01 history edited Connor Mooney CC BY-SA 4.0
added 1 character in body
Aug 14, 2023 at 16:53 comment added Jochen Glueck +1 A small remark: I think one does not only need smoothness of $\partial \Omega$ but also elliptic boundary regularity to get that $u$ and $u_0$ are $C^1$ up to the boundary (which is needed to apply Hopf's lemma and also to get that $u_0$ is dominated by a multiple of $u$, which is part of the first step in your argument). This might somewhat restrict the boundary conditions one can allow for (but I'm not sure for precisely which boundary conditions elliptic boundary conditions still holds or goes wrong).
Aug 14, 2023 at 15:13 history edited Connor Mooney CC BY-SA 4.0
deleted 6 characters in body
Aug 14, 2023 at 10:23 comment added Connor Mooney @Giorgio: thanks for your comment. Yes, I've clarified in my answer that the eigenvalue associated to $u_0$ satisfies $\lambda_0 \leq \lambda$, hence $L(u-u_0) = -\lambda u + \lambda_0 u_0 \leq 0$.
Aug 14, 2023 at 10:21 history edited Connor Mooney CC BY-SA 4.0
added 31 characters in body
Aug 14, 2023 at 8:14 comment added Giorgio Metafune Does your argument work if $u, u_0$ are eigenfunctions with respect to different eigenvalues?
Aug 14, 2023 at 2:57 history edited Connor Mooney CC BY-SA 4.0
deleted 9 characters in body
Aug 13, 2023 at 23:59 history answered Connor Mooney CC BY-SA 4.0