1
$\begingroup$

Let $\Omega$ be a bounded domain in $\mathbb{R}^d$, $d\ge 3$. We define the Neumann-Poincare operator(or double layer potential) $K: L^2(\partial\Omega)\to L^2(\partial\Omega)$ by $$(Kf)(x)=\int_{\partial \Omega}f(y)\frac{\partial}{\partial n_y}E(x,y)dS_y,$$ where $E(x,y)=||x-y||^{2-d}$ and $\frac{\partial}{\partial n_y}$ means the outer normal derivative on the boundary $\partial\Omega$.

Show that $K$ is in the Schatten class $S_p$, $p>d-1$.

Remark: (1) In the paper(http://www.math.ucsb.edu/~mputinar/poincare.pdf), on p.18 the authors only briefly mentioned the above result and a reference book(O.D. Kellogg, Foundations of Potential Theory, J. Springer, Berlin, 1929).

(2) In the paper(https://arxiv.org/pdf/1501.03627.pdf), on p.14 the authors used a kernel condition(Theorem 2.14) for the Schatten class to solve the case $d=3$. But it seems that this condition is not enough to solve higher dimensional cases.

$\endgroup$

1 Answer 1

3
$\begingroup$

If the boundary is Lipschitz, one can prove only that the operator is bounded, it is not necessarily even compact. If the surface is smooth the operator is in fact a pseudodifferential operator of order $-1$, and Weyl asymptotics gives the answer. The latter result holds if the boundary is just a tiny little bit better than Lipschitz, but a hard analysis and perturbation theory is neded.

$\endgroup$
1
  • 1
    $\begingroup$ In the paper ([1], page 15) cited above the boundary is assumed to be of class at least $C^2$, so it is presumable that the OP simply has forgotten to add this condition in his post. $\endgroup$
    – Alex M.
    Commented Jan 26, 2018 at 18:15

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .