Homology boundary links and the Andrews-Curtis conjecture

This paper is concerned with links of n-spheres in \(S^{n+2}\). The main result is: Each sublink of a homology boundary link is concordant to a fusion of a boundary link. The importance of this result comes from the fact that boundary links up to concordance are well-understood. Moreover, recent results of \textit{J. P. Levine}, \textit{W. Mio} and \textit{K. E. Orr} [to appear] and \textit{J. P. Levine} [Invent. Math. 96, 571-592 (1989; Zbl 0692.57010)] show that sublinks of homology boundary links are (up to concordance) characterized by the vanishing of homotopy obstructions. The result above states that the gap between the two classes is ``bridged by a natural band-attaching construction. The authors prove that dropping the concordance condition in the main result is naturally related to the Andrews Curtis Conjecture: If L is a homology boundary link with free group, then L is fusion of a boundary link if and only if its pattern is invertible. Here the pattern is a precise obstruction for a homology boundary link to be a boundary link. A weakening of the AC-conjecture is that all patterns are invertible.

Recommendations

Cited in

(11)

This page was built for publication: Homology boundary links and the Andrews-Curtis conjecture

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q810949)