A Bennequin-type inequality and combinatorial bounds (Q2286442)

From MaRDI portal
Revision as of 10:48, 29 February 2024 by SwMATHimport240215 (talk | contribs) (‎Changed an Item)
scientific article
Language Label Description Also known as
English
A Bennequin-type inequality and combinatorial bounds
scientific article

    Statements

    A Bennequin-type inequality and combinatorial bounds (English)
    0 references
    0 references
    22 January 2020
    0 references
    A contact 3-manifold is a three-dimensional smooth manifold \(M\) endowed with a totally non-integrable plane distribution \(\xi\). A smooth embedding of a number of copies of \(\mathbb S^1\) into \(M\) is called a link in \(M\). A link is called transverse with respect to \(\xi\) if it is nowhere tangent to \(\xi\). Two transverse links in \((M,\xi)\) are equivalent if they are ambient isotopic through a one-parameter family of transverse links. Transverse links are a central object of study in low-dimensional contact topology. Transverse links have two classical invariants, the smooth link-type and a numerical invariant of the equivalence class of a transverse link, the self-linking number \(\mathsf{sl}\). In [Math. Res. Lett. 13, No. 4, 571--586 (2006; Zbl 1143.57006)], \textit{O. Plamenevskaya} defined an invariant of a transverse link in the standard homology sphere, taking values in the Khovanov homology of the link. For a braid \(B\), its Plamenevskaya invariant is a homology class \([\psi(B)]\), of bi-degree \((0,\mathsf{sl}(B))\), in the Khovanov homology of \(\widehat B\), where \(\mathsf{sl}(B)=w(B)-b(B)\) is the self-linking number of a braid \(B\), \(w\) denotes the writhe, i.e., the number of positive crossings minus the number of negative crossings, and \(b\) is the braid index. \textit{J. Rasmussen}, in [Invent. Math. 182, No. 2, 419--447 (2010; Zbl 1211.57009)], introduced an even-valued knot invariant, now widely known as the \(s\)-invariant. It shares some of the basic properties of the knot signature. In particular, it is a concordance invariant and gives the estimation \(s(K)/2\le g_4(K)\) for any knot \(K\), where \(g_4(K)\) is the 4-ball genus. This lower bound for \(g_4(K)\) is powerful and it gives \(g_4(T_{p,q})=u(T_{p,q})=\dfrac{(|p|-1)(|q|-1)}{2}\), where \(u(T_{p,q})\) is the unknotting number of a torus knot of type \((p,q)\). In [Can. J. Math. 60, No. 6, 1240--1266 (2008; Zbl 1171.57010)], \textit{A. Beliakova} and \textit{S. M. Wehrli} generalized the \(s\)-invariant to links as an integer-valued invariant. Some properties of the \(s\)-invariant are generalized. For example, it is a concordance invariant. In [J. Knot Theory Ramifications 28, No. 1, Article ID 1950012, 37 p. (2019; Zbl 1409.57004)], the present author introduced two numerical invariants, called \(c\)-invariants, for transverse links presented as closed braids. These invariants, denoted by \(c_{\mathbb{F}}\) and \(\overline c_{\mathbb{F}}\), are two nonnegative integers. If \(D\) is an oriented link diagram, then it is possible to associate with \(D\) a graph \(\Gamma(D)\), called the simplified Seifert graph. The vertices of \(\Gamma(D)\) are the circles in the oriented resolution of \(D\), and there is an edge between two vertices if the corresponding circles share at least a crossing in \(D\). The edges of the simplified Seifert graph can be divided into three classes (i) positive if the circles corresponding to the endpoints of the edge share only positive crossings, (ii) negative if the circles corresponding to the endpoints of the edge share only negative crossings, and (iii) neutral if the edge is neither positive nor negative. A vertex \(v\) of \(\Gamma(D)\) is called positive, (respectively negative) if there are only positive (respectively negative) edges incident in \(v\). A vertex that is neither positive nor negative is called neutral. Each vertex that is connected only to vertices of the same type as itself is called a pure vertex. If \(D\) is an oriented link diagram and \(\Gamma(D)\) is the simplified Seifert graph of \(D\), then \(V(D)\) is the number of vertices in \(\Gamma(D)\), \(V_{\pm}(D)\) is the number of positive (respectively negative) vertices in \(\Gamma(D)\), \(\ell_{\pm}(D)\) is the number of connected components of \(\Gamma(D)\) composed only of positive (respectively negative) vertices, \(s_{\pm}(D)\) is the number of connected components of the graph obtained from \(\Gamma(D)\) by removing all the negative (respectively positive) edges, and \(\delta_{\_}(D)=1\) if there is a negative edge between two neutral vertices in \(\Gamma(D)\), otherwise \(\delta_{\_}(D)=0\). In this paper, the author provides a new Bennequin-type inequality for the Rasmussen-Beliakova-Wehrli invariant, featuring the numerical transverse braid \(c\)-invariants introduced by the author. From the Bennequin type-inequality and a combinatorial bound on the value of the \(c\)-invariants the author deduces a new computable bound on the Rasmussen invariant. First, it shown that if \(B\) is a braid and \(\mathbb{F}\) is any field, then \[ \overline c_{\mathbb{F}}, c_{\mathbb{F}}\ge V_{\_}(B)-\ell_{\_}(B)+\delta_{\_}(B), \] where \(\overline c_{\mathbb{F}}\) and \(c_{\mathbb{F}}\) are the \(c\)-invariants of the braid \(B\) over the field \(\mathbb{F}\). The second result of this paper presents the inequality from which an upper bound for the \(c\)-invariants can be deduced. If \(\lambda\) is an oriented link type, and \(T\) is a transverse representative of \(\lambda\), then \[ s(\lambda;\mathbb{F})\ge\mathsf{sl}(T)+2c_{\mathbb{F}}(T)+1, \] where \(s(\lambda;\mathbb{F})\) is the Rasmussen-Beliakova-Werhli \(s\)-invariant of \(\lambda\). This inequality is part of a family of inequalities relating topological and transverse invariants. These bounds are named collectively Bennequin-type inequalities. In particular, this inequality sharpens a similar result \(s(\lambda;\mathbb{F})\ge\mathsf{sl}(T)+1\) due to, independently, \textit{O. Plamenevskaya} [Math. Res. Lett. 13, No. 4, 571--586 (2006; Zbl 1143.57006)] and \textit{A. N. Shumakovitch} [Fundam. Math. 225, 343--364 (2014; Zbl 1297.57022)]. Finally, the author shows that if \(D\) is an oriented link diagram representing the oriented link \(L\), then \[ s(\lambda;\mathbb{F})\ge w(D)-V(D)+2V_{\_}(D)-2\ell_{\_}(D)+2\delta_{\_}(D)+1, \] where \(w(D)\) indicates the writhe of \(D\). Moreover, the bound is sharp for negative and positive diagrams.
    0 references
    0 references
    Khovanov homology
    0 references
    Bennequin-type inequality
    0 references
    Rasmussen-Beliakova-Wehrli invariant
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references