Results about the Alexandroff duplicate space (Q2825445)

The Alexandrov duplicate of a topological space is a classical construction in point-set topology. Let \(X\) be a topological space and let \(X^\prime=X\times \{1\}\). Set \(A(X) = X \cup X^\prime\). Then \(A(X)\), called the Alexandrov duplicate of \(X\), may be topologised as follows. Declare every element of \(X^\prime\) isolated. For \(x\in X\) and an open neighbourhood \(U\) of \(x\) in \(X\) let \(U^\prime = U \cup \big((U\times \{1\}) \setminus \{(x,1)\}\big)\). Then we declare \(U^\prime\) an open neighbourhood of \(x\) in \(A(X)\) and, thus, we have introduced a topology in \(A(X)\) by specifying a basis of neighbourhoods of every point in it. This construction was originally performed by Alexandrov in the case where \(X\) is the unit circle \(\mathbb{T}\) (hence the name; sometimes \(A(\mathbb{T}\)) is called the double circle space) and generalised to arbitrary spaces by [\textit{R. Engelking}, Bull. Acad. Pol. Sci., Sér. Sci. Math. Astron. Phys. 16, 629--634 (1968; Zbl 0167.21001)]; see also [\textit{R. E. Chandler} et al., Proc. Am. Math. Soc. 83, 606--608 (1981; Zbl 0473.54015)].NEWLINENEWLINEIt is easily seen that the Alexandrov duplicate of a compact space is compact, too, but metrisability is quickly lost (for example, the Alexandrov duplicate of the circle \(\mathbb{T}\) is already non-metrisable as it contains an uncountable, discrete subspace; namely \(\mathbb{T}^\prime\)).NEWLINENEWLINEThe paper under review studies topological properties that are preserved (or not preserved) by performing the construction of the Alexandrov duplicate. It is proved that if \(X\) is \(\alpha\)-normal, then so is \(A(X)\) (a space \(X\) is \(\alpha\)-normal whenever for every pair of disjoint closed subsets \(A,B\) of \(X\) there exist disjoint open sets \(U,V\) in \(X\) such that \(A\cap U\) is dense in \(A\) and \(B\cap V\) is dense in \(V\)). It is then remarked that extremal disconnectedness is not preserved by this operation. Moreover, a space \(X\) is countably compact if and only if \(A(X)\) is pseudocompact, so pseudocompactness is not preserved by the operation of taking the Alexandrov duplicate either. On the positive side, the authors prove that if \(X\) is epinormal, then so is \(A(X)\) (we refer the reader to the paper for the exact definition of epinormality). The author leave the following two questions open: Is normality preserved by taking the Alexandrov duplicate? Is \(\beta\)-normality preserved by taking the Alexandrov duplicate?