An example on \(s\)-\(H\)-convexity in \(\mathbb{C}^2\) (Q2659505): Difference between revisions

\textit{J. Chaumat} and \textit{A.-M. Chollet} [C. R. Acad. Sci., Paris, Sér. I 306, No. 14, 585--588 (1988; Zbl 0658.35067)] introduced the notion of \(\mathit{s}\)-\(\mathrm{H}\)-convexity. Namely, for any given \(s \in \mathbb{R}\) with \(s \ge 1\), a non-empty compact set \(K \subset {\mathbb C}^n\) is called \(\mathit{s}\)-\(\mathrm{H}\)-convex if there exists a \( C\) with \(0 < C \le 1\) such that, for all \(\epsilon\) with \( 0 < \epsilon \le 1\), there exists an open pseudoconvex subset \(\Omega_{\epsilon}\) of \(\mathbb{C}^n\) satisfying \[\big\{z \in \mathbb{C}^n: d(z, K) < C \epsilon^s\big\} \subset \Omega_{\epsilon}\subset \big\{z \in \mathbb{C}^n: d(z, K) < \epsilon\big\}, \] where \(d(\cdot, K)\) denotes the Euclidean distance to \(K\). Furthermore it is known that there exists a \(k_0(s,n)> 0\) such that \(\mathcal O(\bar\Omega)\) is dense in \(C^k(\bar\Omega) \cap \mathcal O(\Omega)\) for some \(k \ge k_0(s,n)\) and \(\Omega \subset \mathbb{C}^n\) is a bounded pseudoconvex domain, satisfying suitable assumptions, whose closure is \(\mathit{s}\)-\(\mathrm{H}\)-convex (Mergelyan property). So the question is: Given a bounded pseudoconvex domain in \(\mathbb{C}^n\) whose closure admits a Stein neighborhood basis, which additional assumptions guarantee that the closure is \(\mathit{s}\)-\(\mathrm{H}\)-convex, for some \(s \ge 1\)? The main result of this paper is the following: Theorem. There exists a bounded pseudoconvex domain in \(\Omega\subset \mathbb{C}^2\) with boundary of class \(C^{1,1}\) such that: a) \(\bar\Omega\) has a Stein neighborhood basis, b) \(\bar\Omega\) is not \(\mathit{s}\)-\(\mathrm{H}\)-convex for any real number \(s \ge 1\). The proof uses some modifications of the construction of the classical Diederich-Fornaess worm domain [\textit{K. Diederich} and \textit{J. E. Fornaess}, Math. Ann. 225, 275--292 (1977; Zbl 0327.32008)].