A refined Jones polynomial for symmetric unions (Q661389): Difference between revisions

Let \(K_+\) denote a knot and \(K_-\) its mirror image. A knot diagram \(D\) is defined to be a symmetric union of \(K_+\) and \(K_-\) if it is obtained from a connected sum of \(K_+\sharp K_-\) by inserting an arbitrary number of crossings on the chosen symmetry axis. It is well known (see, for example, \textit{R. H. Fox} and \textit{J. W. Milnor}, [Osaka J. Math. 3, 257--267 (1966; Zbl 0146.45501)]) that each symmetric union represents a ribbon knot, but the converse question is still open. The authors of the given paper study the symmetric union in a systematic way. The problem of existence of a representation of a knot as a symmetric union and the question of uniqueness of such a representation are treated in the paper. An appropriate version of the Reidemeister moves (called the symmetric Reidemeister moves) is introduced and studied by the authors. Moreover the authors construct a two-variable refinement \(W_D(s,t)\) of the Jones polynomial that is invariant under symmetric Reidemeister moves, where the variables \(s\) and \(t\) are associated to the two types of crossings on a diagram (on and off the symmetry axis). Notice that the polynomial \(W_D(s,t)\) reflects some geometric properties of ribbon knots represented by symmetric unions (see Corollary 5.3). It also elucidates the connection between the Jones polynomial of the knot \(K\) and its partial knots \(K_+\) and \(K_-\), i.e. it satisfies equalities \(W_D(t,t)= V_K(t)\) and \(W_D(-1,t)=V_{K_-}\cdot V_{K_+}\). The general properties of the \(W\)-polynomial and its behavior on symmetric unions are described in the paper. It turns out that many ribbon knots have essentially distinct symmetric union representations. In particular, the authors prove this for the knot \(9_{27}\) and the two-bridge knots of the form \(K(a,b)=C(2a, 2, 2b, -2, -2a, 2b)\) with \(b=\pm 1\), defined previously in [\textit{M. Eisermann} and \textit{C. Lamm}, J. Knot Theory Ramifications 16, No. 7, 879--898 (2007; Zbl 1157.57004)]. This is done in the paper under review by using the \(W\)-polynomial.