On the nonexistence of parabolic boundary points of certain domains in \(\mathbb C^2\) (Q764937): Difference between revisions

Given a domain \(\Omega\in\mathbb C^2\), a boundary point \(p\in\partial\Omega\) is said to be parabolic if we have \(p=\lim_{t\to\pm\infty} h^t(z)\) \((z\in\Omega)\) for some one-parameter group \(\{ h^t\}_{t\in\mathbb R} \subset \text{Aut}(\Omega)\) of holomorphic automorphisms. The authors show that, under mild additional conditions, there are no boundary parabolic elements in the set \(P_\infty(\partial\Omega)\) of all boundary points of infinite type whenever the latter is a closed curve. In particular they establish that \((0,0)\in P_\infty(\partial\Omega)\) is not of parabolic type provided \(\partial\Omega\) is \(C^\infty\)-smooth satisfying Bell's condition R and there is a neighborhood \(U\) of \((0,0)\) such that \[ \Omega\cap U = \big\{ (z_1,z_2):\;\text{Re\,}z_1 + P(z_2)+| z|^4 | \text{Im} z_1| R(z_2,\text{Im}z_1) \big\} \] with suitable smooth functions \(P,R\), where \(P\) is subharmonic and strictly positive outside the origin where it vanishes to any order.