On 2-universal knots (Q2493409): Difference between revisions

A knot or link \(K\) is \textit{universal} if every closed orientable 3-manifold \(M\) occurs as a finite branched covering of \(S^3\) with branch set \(K\); the authors call a knot or link \textit{2-universal} if, in addition, the ramification index of each component of the preimage of \(K\) in \(M\) can be chosen to be equal to one or two. Examples of 2-universal links were known before, see \textit{M. T. Lozano} and \textit{J. M. Montesinos-Amilibia} [Pac. J. Math. 177, 109--147 (1997; Zbl 0887.58044)], and the main result of the present paper is the existence of 2-universal knots (by a constructive proof, but the authors ask which of the tabulated low-crossing knots are 2-universal). The question is then raised whether there exist 2-universal knots which are \(\pi\)-hyperbolic (i.e., \((S^3,K)\) is a hyperbolic orbifold or cone-manifold, with cone angles of 180 degrees around the singular set \(K\)). This in turn would imply the existence of a universal group \(G\) of isometries of hyperbolic 3-space \(\mathbb H^3\) whose torsion consists only of one conjugacy class of involutions; \(G\) is the universal covering group of the orbifold \((S^3,K)\) (or the ``\(\pi\)-orbifold group'' of the knot, as advertised in a paper by \textit{M. Boileau} and the reviewer [Math. Z. 200, 187--208 (1989; Zbl 0663.57006)]. Recall that a group \(G\) of hyperbolic isometries is universal if every closed orientable 3-manifold occurs as the quotient of \(\mathbb H^3\) by a subgroup of finite index of \(G\); so there remains the question whether there exist \(\pi\)-hyperbolic knots whose \(\pi\)-orbifold groups are universal. For example, 2-bridge knots with cone angles of 180 degrees give spherical orbifolds and are not 2-universal (in fact, their \(\pi\)-orbifold groups are dihedral).