Let $\Omega$ an open bounded domain in $R^n$. Let $u,v$ be nonconstant smooth functions in the interior of $\Omega$ and harmonic in $\Omega$. Suppose that $u,v \in C(\overline{\Omega})$ and $u \geq v$ in $\Omega$. Suppose that exists $x_0 \in \partial \Omega$ with $u(x_0) = v(x_0)$ and $|\nabla u| \leq 1$ in $\Omega$. (I took $1$ for convenience). Is the following inequality true?

$$ \limsup_{y \rightarrow x_0} |\nabla u (y)| \geq \limsup_{y \rightarrow x_0} |\nabla v (y)| $$

I dont know if is true, but if yes, it will help me to understand a passage of a paper. I tried to prove, but no success. Someone could give me a help?

thanks in advance!

This question is related to the last inequality of the case i) (and the case ii)) of lemma 2.3 of the article Existence of classical solutions to a free boundary problem for the p-Laplace operator, (I) by A. Henrot and H. Shahgholian

(the article is in the right side of the page)

  • $\begingroup$ Is it possible in these assumptions that $u$ is constant? Just trying to plausibility check this in my head. $\endgroup$ Jul 31 '14 at 20:00
  • $\begingroup$ ops, assume that the functions is non constant . I forget to write this. sorry $\endgroup$ Jul 31 '14 at 20:05
  • $\begingroup$ If we change $u,v$ with $-u,-v$ the inequality changes "direction" so if it is true it implies equality $\endgroup$
    – user126154
    Jul 31 '14 at 20:45
  • $\begingroup$ Saying it's about "a passage of a paper" but not telling us which one is not so helpful. $\endgroup$
    – user147263
    Jul 31 '14 at 20:52
  • 1
    $\begingroup$ I'll elaborate on the previous comment. You see claim X in a paper. You think that it may be true because Y holds. So you ask: is Y true? Others reply: no, it isn't. Everyone's time has been wasted. Just ask about X. $\endgroup$
    – user147263
    Jul 31 '14 at 21:03

No, this is not true. Let $\Omega$ be unit disk in the plane, and let $u(x_1,x_2) = x_1$. Define $v$ on the boundary so that $v(x_1,x_2) = x_1 - M|x_2|$, where $M$ is large. Then $\partial v/\partial \theta$ is the harmonic function which takes large values on the boundary near $(1,0)$, approximately $\pm M$. Therefore, it will take on large values near $(1,0)$ inside the domain.

  • $\begingroup$ i updated the question $\endgroup$ Jul 31 '14 at 21:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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