On roots of the \(\xi\)-polynomial (Q997800)

For virtual links, \textit{V. O. Manturov} [Russ. Math. Surv. 57, No. 5, 997--998 (2002); translation from Usp. Mat. Nauk 57, No. 5, 141--142 (2002; Zbl 1044.57003)] introduced a two-variable polynomial invariant, called the \(\xi\)-polynomial with variables \(s\) and \(t\). It is defined via an Alexander-like matrix for a planar diagram of a virtual link. See also the same author's paper [J. Knot Theory Ramifications 12, No. 8, 1131--1144 (2003; Zbl 1061.57013)], where the \(\xi\)-polynomial is generalized to a multi-variable polynomial, and [Knot theory. Boca Raton, FL: Chapman \& Hall/CRC (2004; Zbl 1052.57001)]. He showed that it vanishes on classical links, and that there exists a certain skein relation. The polynomial can distinguish some virtual links that cannot be recognized by the Jones polynomial. The paper under review proves that \(t=1\) is a root of the \(\xi\)-polynomial for each virtual knot. This fact does not hold for virtual links, but it does if a link diagram contains one component that has no virtual crossings with the other components.