Semiclassical parametrix for the Maxwell equation and applications to the electromagnetic transmission eigenvalues (Q2036399)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Semiclassical parametrix for the Maxwell equation and applications to the electromagnetic transmission eigenvalues
scientific article

    Statements

    Semiclassical parametrix for the Maxwell equation and applications to the electromagnetic transmission eigenvalues (English)
    0 references
    29 June 2021
    0 references
    If we consider the spectral problem for the Maxwell operator \[A_{\mathrm{Max}}=\frac{1}{\mathrm{i}}\left(\begin{array}{cc} 0 & -\epsilon^{-1}\mathrm{curl}\\ \mu^{-1}\mathrm{curl} & 0 \end{array}\right),\] \(\epsilon,\mu\) strictly positive, bounded, smooth, scalar functions, which is symmetric on smooth electromagnetic fields with compact support in a bounded domain \(\Omega\) with respect to a suitably weighted \(L^{2}\left(\Omega\right)\)-type norm, we have as two extreme cases: boundary conditions ensuring selfadjointness are imposed on the electric or the magnetic part only. In the Maxwell case these two boundary conditions are the same, namely \(n\times E=0\) and \(n\times H=0\) (\(n\) normal vector) on the smooth boundary \(\partial\Omega\), respectively. Compared to an analogous first order formulation of acoustics they would, however, correspond to Dirichlet and Neumann boundary condition, respectively. With inhomogeneous boundary data one can define in analogy to the acoustic case a Dirichlet-to-Neumann operator which associates the boundary data of the electric field with the boundary data of the magnetic field: \(n\times E\mapsto n\times H.\) For this to be a bijection, we need to stay away from the spectrum of the corresponding selfadjoint cases, for example \(\Im\lambda\not=0\), which in the present paper is a constraint somewhat hidden in later parts. The Dirichlet-to-Neumann operator is analyzed by constructing a parametrix near the boundary and calculating its principle symbol. As an application the case of so-called transmission eigenvalue problems is considered. These are eigenvalue problems of a coupled system of two Maxwell operators (labelled by \(1\) and \(2\), respectively) with different speeds of light, \(\epsilon_{1}\mu_{1}\not=\epsilon_{2}\mu_{2}\), in the same domain \(\Omega\) coupled on the boundary \(\partial\Omega\) via \(n\times E_{1}=n\times E_{2}\) and \(n\times c_{1}H_{1}=n\times c_{2}H_{2}\) with \(\frac{c_{1}}{\mu_{1}}=\frac{c_{2}}{\mu_{2}}\). The parametrix construction is applied to allocate a parabolic region without transmission eigenvalues.
    0 references
    0 references
    Maxwell equation
    0 references
    semiclassical parametrix
    0 references
    transmission eigenvalues
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references