I asked a similar question the other day, but I will be more precise now.
Consider $ \Omega:=(0,1 ) \times (0,1)$ and consider
$$ - u_{xx}(x,y) - u_{yy}(x,y) + a(x) u_x + b(y) u_y + u = f(x,y) \mbox{ in } \Omega $$ with $ \partial_\nu u=0$ on $\partial \Omega$ (assume one has some sort of variational solution $u \in H^1(\Omega)$. Suppose $ a(x),b(y)$ are smooth in $ \Omega$ and assume $f$ is Holder continuous on $ \Omega$. \

By looking at the even extension of $u$ (across the various boundaries and which then makes one look at odd extensions of $a$ and $b$ which have jump discontinuities) i seem to be able to prove that $ u \in C^{2,\alpha}( \overline{\Omega})$. I am curious whether this result is true or ? So my question is: does this regularity seem correct or is it known one does not have this regularity at the corners? (the result suprised me a bit but I am also not familiar with what to expect with Neumann boundary conditions).
thanks a bunch for any comments.


1 Answer 1


With appropriate regularity conditions, this is the conclusion of Theorem in:

Grisvard, P.(F-NICE) Elliptic problems in nonsmooth domains. Monographs and Studies in Mathematics, 24.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.