\( \mathbb{Z} \)-disks in \({\mathbb{C}} P^2\) (Q6936217)

Given a closed 4-manifold \(X\), a \(\mathbb{Z}\)-disk in \(X\) is a properly embedded locally flat disk in \(X^\circ := X \setminus B^4\) (so that its boundary is some knot \(K \subset \partial X^\circ \cong S^3\)) whose complement has fundamental group \(\mathbb{Z}\). \textit{M. H. Freedman} and \textit{F. S. Quinn} proved that a knot \(K \subset S^3\) bounds a \(\mathbb{Z}\)-disk in \(S^4\) (which is a slice disk for \(K\) since \((S^4)^\circ \cong B^4\)) if and only if it has Alexander polynomial \(\Delta_K(t) \equiv 1\) (up to multiplication by \(\pm t^k\), \(k\in\mathbb{Z}\), as usual for Alexander polynomials) [Theorem 11.7B in Topology of 4-manifolds. Princeton, NJ: Princeton University Press (1990; Zbl 0705.57001)]. This paper studies \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\), and the main result says that \(\mathbb{Z}\)-disks with boundary \(K\) in \(\mathbb{C}P^2\) necessarily arise by performing a single positive crossing change to a knot with Alexander polynomial 1 and capping off with the \(\mathbb{Z}\)-disk from Freedman's result. Moreover, if the crossing change results in an unknot, then the \(\mathbb{Z}\)-disk can be smoothly realized. Furthermore, for any fixed knot, the authors give an algebraic classification of \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\) with boundary \(K\) up to topological isotopy rel boundary in terms of the behavior of certain elements of the Alexander module under the Blanchfield form. With some number theory, this allows them to say for any knot \(K\) with Alexander polynomial of degree at most 2 exactly how many \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\) the knot \(K\) bounds up to topological isotopy rel boundary; and in particular the number is 1, 2, 4, or infinitely many depending only on the Alexander polynomial.\N\NThe authors include examples with explicit realizations of the \(\mathbb{Z}\)-slice disks in \(\mathbb{C}P^2\) for knots with 1, 2, and 4 such disks up to topological isotopy rel boundary. Furthermore, they give smooth realizations for some of their examples, and note that some of their other examples may or may not have smooth realizations.\N\NThere are many interesting question in 4-dimensional topology related to this paper. For example, are any of the authors' examples nonsmoothable? Is there a genus one knot with infinitely many smooth \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\) which are not even topologically isotopic rel boundary? Can one find exotic \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\) using their results to guarantee topological isotopy rel boundary (for example, by producing smoothly nonisotopic \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\) bounded by a knot which their results guarantee has only one \(\mathbb{Z}\)-disk up to topological isotopy)?\N\NThis work expands on work of \textit{M. E. Fogel} [Pac. J. Math. 163, No. 2, 277--295 (1994; Zbl 0809.57004)] and \textit{M. Borodzik} and \textit{S. Friedl} [Trans. Lond. Math. Soc. 1, No. 1, 57--84 (2014; Zbl 1322.57010)], who give similar algebraic criteria for the existence of \(\mathbb{Z}\)-disks in \(\mathbb{C}P^2\). The contribution of this paper is relating the relevant algebraic object explicitly to a geometric construction of the \(\mathbb{Z}\)-disks, which then allows the authors to precisely count the \(\mathbb{Z}\)-disks for any particular knot.