On the Steinness index (Q2421235): Difference between revisions

The notion of ``Steiness index'' is introduced and related to the existence of Stein neighborhood bases. The Steiness index complements the Diederich-Fornæss exponent, introduced by \textit{K. Diederich} and \textit{J. E. Fornaess} [Invent. Math. 39, 129--141 (1977; Zbl 0353.32025)]. Recall that, for a defining function, \(\rho\), of a smoothly bounded domain, the Diederich-Fornæss exponent is defined as \[ \sup\{ \eta \in (0,1) : -(-\rho)^{\eta} \text{ is strictly plurisubharmonic on }\Omega\}. \] The Diedrich-Fornæss index is the supremum of exponents over all defining functions. The existence of the Diederich-Fornæss index implies the existence of a strictly plurisubharmonic exhaustion function. By contrast, the Steiness exponent for \(\rho\) is defined by \[ \inf \{ \eta > 1 : \rho^{\eta} \text{ is strictly plurisubharmonic on }\overline{\Omega}^c \text{ in a neighborhood of }\partial\Omega\}, \] while the Steiness index is defined as the infimum of Steiness exponents over all \(\rho\). The main result of the article is that the existence of a (finite) Steiness index implies the existence of a strong Stein neighborhood basis (a defining function for \(\Omega\) exists whose level sets, \(\rho=\epsilon\), define pseudoconvex boundaries for all small positive \(\epsilon\)). The main idea in the proof is the relation of the Steiness index to inequalities involving the Levi form on vector fields at weakly pseudoconvex boundary points. It is further shown that it is sufficient to consider boundary points of infinite type. Furthermore, some explicit calculations of the Steiness index on the Diederich-Fornæss worm domains and on convex domains are also given.